Comparison of Procedure-Based and Diagnosis-Based Identifications of Severe Sepsis and Disseminated Intravascular Coagulation in Administrative Data

Studies of Health Insurance Claims Data in Japan: A Scoping Review

BACKGROUND

Health insurance claims data are used in various research fields; however, an overview on how they are used in healthcare research is scarce in Japan. Therefore, we conducted a scoping review to systematically map the relevant studies using Japanese claims data.

METHODS

MEDLINE, EMBASE, and Ichushi-Web were searched up to April 2021 for studies using Japanese healthcare claims data. We abstracted the data on study characteristics and summarized target diseases and research themes by the types of claims database. Moreover, we described the results of studies that aimed to compare health insurance claims data with other data sources narratively.

RESULTS

A total of 1,493 studies were included. Overall, the most common disease classifications were "Diseases of the circulatory system" (18.8%, n = 281), "Endocrine, nutritional, and metabolic diseases" (11.5%, n = 171; mostly diabetes), and "Neoplasms" (10.9%, n = 162), and the most common research themes were "medical treatment status" (30.0%, n = 448), "intervention effect" (29.9%, n = 447), and "clinical epidemiology, course of diseases" (27.9%, n = 417). Frequent diseases and themes varied by type of claims databases. A total of 19 studies aimed to assess the validity of the claims-based definition, and 21 aimed to compare the results of claims data with other data sources. Most studies that assessed the validity of claims data compared to medical records were hospital-based, with a small number of institutions.

CONCLUSIONS

Claims data are used in various research areas and will increasingly provide important evidence for healthcare policy in Japan. It is important to use previous claims database studies and share information on methodology among researchers, including validation studies, while informing policymakers about the applicability of claims data for healthcare planning and management.

Validation studies of Japanese administrative health care data: A scoping review

PURPOSE

Large-scale administrative health care databases are increasingly being utilized for research. However, there has not been much literature that validated administrative data in Japan; a previous review identified six validation studies published between 2011 and 2017. We conducted a literature review of studies that assessed the validity of Japanese administrative health care data.

METHODS

We searched for studies published by March 2022 that compared individual-level administrative data with a reference standard from another data source, as well as studies that validated administrative data using other data within the same database. The eligible studies were also summarized based on characteristics which included data types, settings, reference standard used, numbers of patients, and conditions validated.

RESULTS

There were 36 eligible studies, including 29 that used external reference standard and seven that validated administrative data using other data within the same database. Chart review was the reference standard in 21 studies (range of the numbers of patients, 72-1674; 11 studies conducted in single institutions and nine studies in 2-5 institutions). Five studies used a disease registry as the reference standard. Diagnoses of cardiovascular diseases, cancer, and diabetes were frequently evaluated.

CONCLUSIONS

Validation studies are being conducted at an increasing rate in Japan, although most of them are small scale. Further large-scale comprehensive validation studies are necessary to effectively utilize the databases for research.

Comparison of bleeding following gastrointestinal endoscopic biopsy in patients treated with and without direct oral anticoagulants

Background and study aims  Despite the widespread use of direct oral anticoagulants (DOACs), the association between DOAC use and complications (e. g., bleeding) following gastrointestinal endoscopic biopsy remains unclear. This study aimed to evaluate complications after biopsy in patients treated with DOACs in Japan, where biopsies would be generally performed without DOAC withdrawal based on guideline recommendations. Patients and methods  Using a Japanese nationwide database, we identified patients taking DOACs who underwent gastrointestinal endoscopic biopsy (n  = 2,769, DOAC group) and those not taking DOACs (n = 129,357, control group) from April 2015 to November 2020. We conducted 1:4 propensity score (PS) matching and overlap PS-weighting analyses with adjustment for background characteristics to compare occurrence of post-procedure hemorrhage and stroke within 1 week after biopsy, and thrombin use on the day of biopsy without a diagnosis of hemorrhage. Results  In total, 578 patients (0.44 %) developed post-procedure hemorrhage, and 13 patients (0.01 %) developed stroke. The DOAC group had more comorbidities than the control group. The PS matching analysis revealed no significant differences in post-procedure hemorrhage (odds ratio, 1.52 [95 % confidential interval, 0.96-2.41]) or stroke (1.00 [0.21-4.71]), whereas the DOAC group received thrombin more often than the control group (1.60 [1.30-1.95]). The results were equivalent in the overlap PS-weighting analysis. Conclusions  The PS analyses showed no significant differences in complications following gastrointestinal endoscopic biopsy between DOAC users and non-users. These results suggest the safety of endoscopic biopsy without DOAC withdrawal although the need for careful hemostasis remains.

Real-world evidence on disseminated intravascular coagulation from Japan

Many descriptive epidemiological and comparative studies using big data have been reported recently from outside Japan. Within Japan, diagnosis procedure combination (DPC) data and medical receipt data are being stored in electronic media, and real-world evidence in various fields has started to be reported. We reviewed clinical studies on disseminated intravascular coagulation (DIC) using DPC data obtained from an insurance database with large numbers of cases and a related commercially available dataset including DPC and laboratory data. After DPC was introduced in 2003, 19 studies on DIC using Japanese national DPC data and two studies using the Medical Data Vision database were reported. Epidemiological findings in seven studies showed that the proportion of drugs administered for each underlying disease differed, with antithrombin and recombinant thrombomodulin (rTM) being used more frequently in clinical settings. In 14 comparative studies on anti-DIC agents, antithrombin for severe pneumonia, postoperative intestinal perforation, and severe burn, and rTM for acute cholangitis were associated with improved survival rates. Large-scale observational studies using big data can show results similar to those of randomized control trials if the quality of individual research is high. Real-world data analysis will be increasingly necessary to complement the evidence gap unfilled by randomized control trials.

Existing Data Sources for Clinical Epidemiology: Database of the National Hospital Organization in Japan

This review introduces the National Hospital Organization (NHO) database in Japan. The NHO has maintained two databases through a system of data collection from 140 hospitals in the NHO. National Hospital Organization Clinical Data Archives (NCDA) is collecting clinical information in real time from the electronic medical records since January 2016, and Medical Information Analysis (MIA) databank is collecting daily insurance claims data since April 2010. The NHO database covers more than 8 million patients in 140 hospitals throughout Japan. The database consists of the information of patient profiles, hospital admission and discharge, diagnosis with ICD-10 codes, text data from medical chart, daily health insurance claims such as medical procedures, medications or surgeries, vital signs and laboratory data, and so on. The NHO database includes a wide variety of diseases and settings, including acute, chronic and intractable diseases, emergency medical services, disaster medicine, response to emerging infectious disease outbreaks, medical care according to health policies such as psychiatry, tuberculosis, or muscular dystrophy, and health systems in sparsely populated non-urban areas. Among several common diseases, the database has representativeness in terms of age distribution compared with the Patient Survey 2017 by the Ministry of Health, Labour and Welfare. Interested researchers can contact (700-dbproject@mail.hosp.go.jp) the NHO database division to obtain more information about the NHO database for utilization.

Validity of operative information in Japanese administrative data: a chart review-based analysis of 1221 cases at a single institution

PURPOSE

To evaluate the validity of operative information recorded in the Diagnosis Procedure Combination (DPC) database, a national inpatient database including administrative claims data.

METHODS

We reviewed the medical charts of 1221 patients who underwent one of six surgeries (breast, esophageal, gastric, thyroid cancer surgery, appendectomy, or inguinal hernia repair) at a surgery department of a university hospital from April 2016 to March 2019. We compared operative information (type, date, laterality of procedure; type of anesthesia; transfusion; and duration of anesthesia) recorded in the DPC database with the information recorded in the medical charts.

RESULTS

The DPC data for type, date, laterality of surgery, and type of anesthesia were accurate in 99% of the reviewed patients. The sensitivity and specificity for identifying whether a patient received a transfusion procedure were 97.5% and 99.6%, respectively. Data regarding the duration of anesthesia in the DPC database were identical to those in medical chart records in 1114 of 1216 cases that received general or spinal anesthesia (91.5%). The duration of anesthesia in the DPC data was 53 min longer on average than the recorded operative time in the medical charts.

CONCLUSION

The DPC database had high validity for operative information.

Additional effect of azithromycin over β-lactam alone for severe community-acquired pneumonia-associated acute respiratory distress syndrome: a retrospective cohort study

Background

Community-acquired pneumonia (CAP) is the most common cause of acute respiratory distress syndrome (ARDS). Although previous studies have suggested that macrolide therapy is beneficial for ARDS, its benefit for severe CAP-associated ARDS remains uncertain. Previous studies were limited in that they had a small sample size and included patients with non-pulmonary ARDS and those with pulmonary ARDS. This study aimed to investigate the additional effect of azithromycin when used with β-lactam compared with the effect of β-lactam alone in mechanically ventilated patients with CAP-associated ARDS.

Methods

We identified mechanically ventilated patients with CAP-associated ARDS between July 2010 and March 2015 using data in the Diagnosis Procedure Combination database, a Japanese nationwide inpatient database. We performed propensity score matching analysis to assess 28-day mortality and in-hospital mortality in mechanically ventilated patients with CAP-associated ARDS who received β-lactam with and without azithromycin within hospital 2 days after admission. The inverse probability of treatment weighting analysis was also conducted.

Results

Eligible patients (n = 1257) were divided into the azithromycin group (n = 226) and the control group (n = 1031). The one-to-four propensity score matching analysis included 139 azithromycin users and 556 non-users. No significant difference was observed between the groups with respect to 28-day mortality (34.5% vs. 37.6%, p = 0.556) or in-hospital mortality (46.0% vs. 49.1%, p = 0.569). The inverse probability of treatment weighting analysis showed similar results.

Conclusions

Compared with treatment with β-lactam alone, treatment with azithromycin plus β-lactam had no significant additional effect on 28-day mortality or in-hospital mortality in mechanically ventilated patients with CAP-associated ARDS. To the best of our knowledge, this study is the first to determine the effect of azithromycin in mechanically ventilated patients with CAP-associated ARDS.

Supplementary Information

The online version contains supplementary material available at 10.1186/s41479-021-00093-8.

Association between intensive care unit admission of a patient and mental disorders in the spouse: a retrospective matched-pair cohort study

Background

Previous prospective studies have suggested that spouses of patients who are admitted to the intensive care unit (ICU) have a high prevalence of mental disorders, termed post-intensive care syndrome-family (PICS-F). However, it remains unclear whether the patient’s ICU admission is associated with the occurrence of mental disorders in the spouse outside of the prospective study setting. We therefore investigated the proportion of ICU patients’ spouses who visited medical facilities for mental disorders and the association between ICU admission of a patient and mental disorders in the spouse using real-world data.

Methods

This was a retrospective matched-pair cohort study using commercially available, routinely collected administrative claims data. As the study population, we identified all married couples (both wife and husband) who were registered in the database from 1 April 2012 to 31 August 2018 using family identification codes. We identified spouses of patients who were admitted to the ICU for more than 2 days as the exposure group and defined the date of admission to the ICU as the index date. We randomly matched four individuals in the non-exposure group with one individual in the exposure group. The primary outcome was any PICS-F–related mental disorder in the spouses within 6 months from the index date. As a sensitivity analysis, we also investigated the proportion and association of individuals (excluding spouses) with a history of mental disorders.

Results

Among 1,082,208 married couples, we identified 8490 spouses of ICU patients, and they were matched with 33,946 individuals. The proportion of any PICS-F–related mental disorder within 6 months from the index date was 12.8% in ICU patients’ spouses and 11.3% in the matched individuals (adjusted odds ratio, 1.29; 95% confidence interval, 1.03–1.42). The sensitivity analysis showed significant associations between ICU admission and spouses’ mental disorders.

Conclusions

Spouses of patients who were admitted to the ICU had a slightly higher risk of mental disorders within 6 months than spouses of patients who were not admitted to the ICU.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40560-021-00583-3.

Claims-based algorithms for common chronic conditions were efficiently constructed using machine learning methods

Identification of medical conditions using claims data is generally conducted with algorithms based on subject-matter knowledge. However, these claims-based algorithms (CBAs) are highly dependent on the knowledge level and not necessarily optimized for target conditions. We investigated whether machine learning methods can supplement researchers' knowledge of target conditions in building CBAs. Retrospective cohort study using a claims database combined with annual health check-up results of employees' health insurance programs for fiscal year 2016-17 in Japan (study population for hypertension, N = 631,289; diabetes, N = 152,368; dyslipidemia, N = 614,434). We constructed CBAs with logistic regression, k-nearest neighbor, support vector machine, penalized logistic regression, tree-based model, and neural network for identifying patients with three common chronic conditions: hypertension, diabetes, and dyslipidemia. We then compared their association measures using a completely hold-out test set (25% of the study population). Among the test cohorts of 157,822, 38,092, and 153,608 enrollees for hypertension, diabetes, and dyslipidemia, 25.4%, 8.4%, and 38.7% of them had a diagnosis of the corresponding condition. The areas under the receiver operating characteristic curve (AUCs) of the logistic regression with/without subject-matter knowledge about the target condition were .923/.921 for hypertension, .957/.938 for diabetes, and .739/.747 for dyslipidemia. The logistic lasso, logistic elastic-net, and tree-based methods yielded AUCs comparable to those of the logistic regression with subject-matter knowledge: .923-.931 for hypertension; .958-.966 for diabetes; .747-.773 for dyslipidemia. We found that machine learning methods can attain AUCs comparable to the conventional knowledge-based method in building CBAs.

Effect of antithrombin III among patients with disseminated intravascular coagulation in obstetrics: a nationwide observational study in Japan

OBJECTIVE

Pregnant women may develop disseminated intravascular coagulation (DIC), possibly resulting in massive maternal haemorrhage and perinatal death. The Japan guideline recommends use of antithrombin III (ATIII) for DIC in obstetrics; however, its effect remains uncertain. The present study aimed to investigate the effect of ATIII for DIC in obstetrics, using a national inpatient database in Japan.

DESIGN

Nationwide observational study.

SETTING

Japan.

POPULATION

We used the Diagnosis Procedure Combination inpatient database to identify patients who delivered at hospital and were diagnosed with DIC from July 2010 to March 2018.

METHODS

Propensity score matching analyses were performed to compare in-hospital maternal mortality and hysterectomy during hospitalisation between users and non-users of ATIII on the day of delivery.

MAIN OUTCOME MEASURES

In-hospital mortality, hysterectomy.

RESULTS

A total of 9920 patients were enrolled, including 4329 patients (44%) who used ATIII and 5511 patients (56%) who did not use ATIII. One-to-one propensity score matching created 3290 pairs. In-hospital maternal mortality did not differ significantly between the propensity-matched groups (0.3% in the ATIII group versus 0.5% in the control group; odds ratio 0.73; 95% CI 0.35-1.54). A significantly lower proportion of patients in the ATIII group, compared with those in the control group, underwent hysterectomy during hospitalisation (5.3% versus 8.7%; absolute risk difference -2.9%; 95% CI -4.2 to -1.6%).

CONCLUSIONS

Although the present study did not show a mortality-reducing effect of ATIII for patients with DIC in obstetrics, it may have clinical benefit in terms of reducing the number of patients undergoing hysterectomy.

TWEETABLE ABSTRACT

This study did not show mortality-reducing effect of antithrombin III for patients with DIC in obstetrics.

ERCP improves mortality in acute biliary pancreatitis without cholangitis

Background and study aims  Acute pancreatitis (AP) is an increasingly common indication for hospitalization in the United States. The necessity for endoscopic retrograde cholangiopancreatography (ERCP) and the timing of ERCP in acute gallstone-related pancreatitis without cholangitis (AGPNC) is controversial. The aim of this study was to evaluate the association of ERCP and its performance during admission with mortality and length of stay (LOS) in patients with AGPNC. Patients and methods  We queried the Nationwide Inpatient Sample (NIS) from 2004 to 2014 to identify all patients with admissions for gallstone AP. We excluded patients with chronic pancreatitis or concurrent cholangitis, and those who were transferred from elsewhere for treatment. Our primary outcome measure was inpatient mortality. Our secondary outcome measure was hospital length of stay (LOS). Results  We identified 491,011 records eligible for analysis. Of the patients, 30.6 % (150,101) had AGPNC. There were 1.34 deaths per 100 admissions in patients with AGPNC. The average LOS was 5.88 (± 6.38) days with a median stay of 4 days (range, 3-7). When adjusted for age, Elixhauser Comorbidity Index, and severe pancreatitis, patients with ERCP during admission were 43 % less likely to die. ERCP performed between Days 3 and 9 of hospitalization resulted in a significant mortality benefit. Among those who had ERCP, a shorter wait time for ERCP was associated with a shorter LOS after adjustment for demographics and severity of illness. Conclusion  ERCP performed during inpatient admission for AGPNC was associated with decreased mortality. These data support early ERCP in patients with acute gallstone pancreatitis without cholangitis.

Underlying Disorders, Clinical Phenotypes, and Treatment Diversity among Patients with Disseminated Intravascular Coagulation

INTRODUCTION

Clinical guidelines state that disseminated intravascular coagulation (DIC) treatment should be based on three clinical phenotypes: the marked bleeding type (e.g. leukemia, trauma, obstetric diseases, or aortic diseases); organ failure type (sepsis or pancreatitis); and asymptomatic type of DIC (solid cancer). However, among the various underlying disorders of DIC, the clinical presentations of bleeding or organ failure have not to date been well documented. The present study aimed to evaluate whether underlying disorders of DIC would affect clinical outcome including death, organ failure, and bleeding.

METHODS

Using the Japanese Diagnosis Procedure Combination inpatient database, we identified all adult patients diagnosed with DIC during hospitalization from July 1, 2010, to March 31, 2018. We collected data on patient characteristics and underlying disorders of DIC including sepsis, solid cancer, leukemia, trauma, obstetric diseases, aortic diseases, pancreatitis, and miscellaneous diseases. We counted major bleeding events and calculated an organ failure score for patients during hospitalization.

RESULTS

We identified 337,132 patients with DIC. The major disorders underlying DIC were sepsis (42%) and solid cancer (31%). The average organ failure scores of patients with aortic diseases, sepsis, and trauma were 2.8, 2.2, and 2.2, respectively. The percentages with major bleeding events among patients with aortic diseases, trauma, obstetric diseases, and solid cancer were 24%, 15%, 10%, and 10%, respectively.

CONCLUSIONS

This study suggests that the clinical presentations of bleeding and organ failure are not associated with the three existing clinical phenotypes of DIC or with the underlying disorders of DIC. Therefore, clinical presentation alone may not be sufficient for identifying the clinical phenotypes of DIC. Further research is necessary to develop new strategies for identifying the phenotypes of DIC and improving treatment strategies for individual patients.

Time Trends of the Outcomes and Treatment Options for Disseminated Intravascular Coagulation: A Nationwide Observational Study in Japan

Introduction

Existing evidence on the mortality time trends of patients with disseminated intravascular coagulation (DIC) is limited, and whether the mortality trend or quality of care of DIC patients has improved remains unknown. This study aimed to investigate the temporal trend in mortality, patient outcomes, and treatment preferences of several anticoagulants in Japan.

Methods

This retrospective observational study used the Japanese Diagnosis Procedure Combination inpatient database, which contains data from more than 1200 acute-care hospitals in Japan. We identified all adult patients that were diagnosed with DIC from July 2010 to March 2018 and sorted them into one of five predefined underlying conditions: sepsis, solid cancer, leukemia, trauma, or obstetric. The data collected as general outcomes were the 28-day mortality and major bleeding events. We also evaluated anticoagulant use for DIC treatment.

Results

A total of 325,327 DIC patients were included in this study. Regarding the baseline characteristics, an increase in median age, worsened comorbid conditions, and higher illness severity were observed over time. The underlying conditions for DIC were largely unchanged. Over the study period, the 28-day mortality for overall DIC patients decreased from 41.8% (95% CI 41.2%-42.3%) to 36.1% (95% CI 35.6%-36.6%), which is a 14% decrease over the 8-year period (P _trend < 0.001). The downward trend in mortality was more evident in patients with sepsis and leukemia (15% and 14% decreases, respectively), whereas no clinically meaningful change in mortality occurred in trauma and obstetrics patients. Over time, major bleeding events modestly increased, and the length of hospital stay decreased. The temporal trend in the treatment preferences of anticoagulants for DIC patients clearly changed over time.

Conclusions

The overall 28-day mortality for DIC patients clearly decreased from 2010 to 2017. The downward trend in mortality might have resulted from the advances made in the fundamental treatment of underlying diseases and from the changes in anti-DIC strategies.

Antithrombin use and mortality in patients with stage IV solid tumor-associated disseminated intravascular coagulation: a nationwide observational study in Japan

Background

Terminal-stage solid tumors are one of the main causes of disseminated intravascular coagulation (DIC); effective therapeutic strategies are therefore warranted. This study aimed to investigate the association between mortality and antithrombin therapy in patients with stage IV solid tumor-associated DIC using a large nationwide inpatient database.

Methods

From July 2010 to March 2018, patients with stage IV solid tumor-associated DIC in the general wards, intensive care unit, or high care unit were identified using the Japanese Diagnosis Procedure Combination Inpatient Database. Patients who received antithrombin within 3 days of admission were allocated to the antithrombin group, while the remaining patients were allocated to the control group. One-to-four propensity score matching analyses were applied to compare outcomes. The primary outcome was the 28-day in-hospital mortality.

Results

Of the 25,299 eligible patients, 919 patients had received antithrombin within 3 days of admission and were matched with 3676 patients in the control group. There were no significant differences in the 28-day mortality between the two groups (control vs. antithrombin: 28.9% vs. 30.3%; hazard ratio, 1.08; 95% confidence interval, 0.95–1.23). There were no significant differences in the organ failure score and the proportion of critical bleeding between the two groups. Subgroup analyses showed that the effects of antithrombin were not significantly different among different tumor types.

Conclusion

Using a nationwide Japanese inpatient database, this study showed that there is no association between antithrombin administration and 28-day mortality in patients with stage IV solid tumor-associated DIC. Therefore, establishing other therapeutic strategies for solid tumor-associated DIC is required.

Recombinant Thrombomodulin in Disseminated Intravascular Coagulation Associated with Stage IV Solid Tumors: A Nationwide Observational Study in Japan

OBJECTIVE

 The terminal stage of solid tumors sometimes induces disseminated intravascular coagulation (DIC); however, no useful therapeutic strategies have been established. This study investigated the relationship between mortality and recombinant human soluble thrombomodulin (rTM) therapy for patients with DIC associated with stage IV solid tumors using a large nationwide inpatient database.

METHODS

 Using the Japanese Diagnosis Procedure Combination Inpatient Database, patients with stage IV solid tumors who developed DIC were identified. Those who received rTM within 3 days of admission were included in the treatment group; the remaining were included in the control group. The primary outcome was the 28-day in-hospital mortality.

RESULTS

 Of 25,299 eligible patients, 1 to 4 propensity score matching was used to select 1,979 rTM users and 7,916 nonusers. There was no significant difference in the 28-day mortality (control vs. rTM: 37.4% vs. 34.3%; hazard ratio, 0.95; 95% confidence interval [CI], 0.88-1.04) and critical bleeding rate (control vs. rTM: 3.7% vs. 3.8%; odds ratio, 1.04; 95% CI, 0.75-1.42) between groups. Subgroup analyses showed that the 28-day mortality rate among patients with colorectal and gynecological cancer was significantly lower in the rTM than in the control group (p for interaction 0.033 and 0.010, respectively).

CONCLUSION

 Although we identified a possibly beneficial association between rTM administration and mortality in specific populations of patients with colorectal and gynecological cancer, no such association was found when considering the entire cohort of patients with DIC associated with stage IV solid tumors.

Histamine-2 receptor antagonists versus proton pump inhibitors for septic shock after lower gastrointestinal tract perforation: a retrospective cohort study using a national inpatient database

Background

Studies have shown the potential benefit of stress ulcer prophylaxis including histamine-2 receptor antagonists (H2RA) and proton pump inhibitors (PPI) in critically ill patients. However, the adverse effects of stress ulcer prophylaxis such as Clostridioides difficile infection (CDI) and hospital-acquired pneumonia have been reported. Abdominal septic shock is associated with increased risk of bleeding, CDI, and pneumonia; however, which ulcer prophylaxis might be associated with better outcomes in patients with septic shock after lower gastrointestinal tract perforation is unknown.

Methods

In this retrospective cohort study using the Japanese Diagnosis Procedure Combination database from July 2010 to March 2015, we identified patients aged 18 years or older who received open abdominal surgery for lower gastrointestinal tract perforation and who used vasopressors and antibiotics within 2 days of admission. We performed propensity score matching and inverse probability of treatment weighting (IPTW) to compare the outcomes between patients who received H2RA and those who received PPI within 2 days of admission. The outcomes included gastrointestinal bleeding requiring endoscopic hemostasis within 28 days of admission, 28-day mortality, CDI, and hospital-acquired pneumonia.

Results

The propensity score matching created 1088 pairs of patients who received H2RA or PPI within 2 days of admission. There were no significant differences between the H2RA and PPI groups regarding gastrointestinal bleeding requiring endoscopic hemostasis within 28 days of admission (0.74% vs 1.3%, risk ratio 0.57 (0.24–1.4), and P = 0.284), 28-day mortality (11.3% vs 12.9%, risk ratio 0.88 (0.68–1.1), and P = 0.386), CDI (0.64% vs 0.46%, risk ratio 1.4 (0.45–4.4), and P = 0.774), and hospital-acquired pneumonia (3.0% vs 4.3%, risk ratio 0.70 (0.45–1.1), and P = 0.138). IPTW analysis showed similar results.

Conclusions

There were no significant differences in gastrointestinal bleeding requiring endoscopic hemostasis within 28 days of admission, 28-day mortality, CDI, and hospital-acquired pneumonia between H2RA and PPI in patients with septic shock after lower gastrointestinal tract perforation.

Levetiracetam vs. Fosphenytoin for Second-Line Treatment of Status Epilepticus: Propensity Score Matching Analysis Using a Nationwide Inpatient Database

Objective: Status epilepticus is a major emergency condition. The choice of antiepileptic drugs for second-line treatment after benzodiazepine remains controversial, including levetiracetam vs. fosphenytoin. We compare the safety of intravenous levetiracetam and fosphenytoin as a second-line treatment in patients with status epilepticus using a nationwide database.

Methods: An observational study conducted with the Japanese Diagnosis Procedure Combination inpatient database identified adult patients who had been admitted for status epilepticus and who had received intravenous diazepam on the day of admission from March 1, 2011 to March 31, 2018. Patients who received intravenous levetiracetam on the day of admission were defined as the levetiracetam group and those who received intravenous fosphenytoin on the day of admission were defined as the fosphenytoin group. Propensity score matching was performed to compare outcomes obtained for the levetiracetam and fosphenytoin groups.

Results: The analysis examined data of 5,667 patients. Overall, 1,403 (25%) patients received levetiracetam; 4,264 (75%) received fosphenytoin. One-to-one propensity score matching created 1,363 matched pairs. No significant difference was found in in-hospital mortality (5.2 vs. 5.1%; odds ratio, 1.03; 95% confidence interval, 0.73–1.46). The proportion of vasopressor use on the day of admission was significantly lower for the levetiracetam group than for the fosphenytoin group (3.2 vs. 4.9%; odds ratio, 0.63; 95% confidence interval, 0.43–0.92). No significant difference was found in other secondary outcomes including total hospitalization cost.

Conclusion: Levetiracetam was related to significantly reduced vasopressor use on the day of admission than that found for fosphenytoin, in adult status epilepticus.

Phenytoin versus fosphenytoin for second-line treatment of status epilepticus: propensity score matching analysis using a nationwide inpatient database

PURPOSE

For status epilepticus, the choice of antiepileptic drugs for second-line treatment after benzodiazepine remains controversial: phenytoin or fosphenytoin are recommended, however, it has been unknown which is better. Using a nationwide database, we compared the efficacy and safety of them.

METHOD

An observational study conducted with the Japanese Diagnosis Procedure Combination inpatient database identified adult patients who had been admitted for status epilepticus and who had received intravenous diazepam on the day of admission from January 1, 2011 through December 31, 2015. Propensity score matching was applied to compare outcomes of the phenytoin and fosphenytoin groups.

RESULTS

The analysis examined data of 5265 patients: 2969 patients received phenytoin; 2296 received fosphenytoin, on the day of admission. One-to-one propensity score matching created 1871 matched pairs. No significant difference was found for vasopressor use on the day of admission (4.2 % vs. 4.4 %; odds ratio 1.07; 95 % confidence intervals 0.77-1.48; p = 0.69), or for mechanical ventilation on the day of admission, in-hospital mortality, length of hospital stay, or total hospitalization cost. Higher age, comorbidity of cardiac diseases and lower body mass index were associated significantly with increased vasopressor use, whereas the dose of phenytoin equivalents and the choice of fosphenytoin were not.

CONCLUSIONS

This nationwide observational study found no evidence that fosphenytoin provides higher efficacy or safety than phenytoin for treatment of status epilepticus in adults after diazepam. Age, cardiac disease and low body mass index were identified as independent risk factors for vasopressor use in both phenytoin and fosphenytoin.

Early enteral nutrition in patients with severe traumatic brain injury: a propensity score-matched analysis using a nationwide inpatient database in Japan

BACKGROUND

Whether enteral nutrition (EN) should be administered early in severe traumatic brain injury (TBI) patients has not been fully addressed.

OBJECTIVE

The present study aimed to evaluate whether early EN can reduce mortality or nosocomial pneumonia among severe TBI patients.

METHODS

Using the Japanese Diagnosis Procedure Combination inpatient database from April 2014 to March 2017 linked with the Survey for Medical Institutions, we identified patients admitted for intracranial injury with Japan Coma Scale scores ≥30 (corresponding to Glasgow Coma Scale scores ≤8) at admission. We designated patients who started EN within 2 d of admission as the early EN group, and those who started EN at 3-5 d after admission as the delayed EN group. The primary outcome was in-hospital mortality. The secondary outcome was nosocomial pneumonia. Propensity score-matched analyses were performed to compare the outcomes between the 2 groups.

RESULTS

We identified 3080 eligible patients during the 36-mo study period, comprising 1100 (36%) in the early EN group and 1980 (64%) in the delayed EN group. After propensity score matching, there was no significant difference in in-hospital mortality (difference: -0.3%; 95% CI: -3.7%, 3.1%) between the 2 groups. The proportion of nosocomial pneumonia was significantly lower in the early EN group than in the delayed EN group (difference: -3.2%; 95% CI: -5.9%, -0.4%).

CONCLUSIONS

Early EN may not reduce mortality, but may reduce nosocomial pneumonia in patients with severe TBI.

The effect of antithrombin added to recombinant human-soluble thrombomodulin for severe community-acquired pneumonia-associated disseminated intravascular coagulation: a retrospective cohort study using a nationwide inpatient database

Background

Studies showed potential benefits of recombinant human-soluble thrombomodulin (rhTM) and antithrombin for treating sepsis associated disseminated intravascular coagulation. However, benefits of their combination have been inconclusive.

Methods

Using a nationwide inpatient database in Japan, we performed propensity-score matched analyses to compare outcomes between rhTM combined with antithrombin and rhTM alone for severe community-acquired pneumonia associated disseminated intravascular coagulation from July 2010 to March 2015. The outcomes included in-hospital mortality and requirement of red cell transfusion.

Results

Propensity score matching created 189 pairs of patients who received rhTM combined with antithrombin or rhTM alone within 2 days of admission. There was no significant difference between the two groups for in-hospital mortality (40.2% vs. 45.5%). Patients treated with rhTM and antithrombin were more likely to require red cell transfusion than those treated with rhTM alone (37.0% vs. 25.9%).

Conclusions

Compared with rhTM alone, combination of rhTM with antithrombin for severe community-acquired pneumonia-associated disseminated intravascular coagulation may be ineffective for reducing mortality and may increase bleeding.

Machine Learning-Based Prediction Models for 30-Day Readmission after Hospitalization for Chronic Obstructive Pulmonary Disease

While machine learning approaches can enhance prediction ability, little is known about their ability to predict 30-day readmission after hospitalization for Chronic Obstructive Pulmonary Disease (COPD). We identified patients aged ≥40 years with unplanned hospitalization due to COPD in the Diagnosis Procedure Combination database, an administrative claims database in Japan, from 2011 through 2016 (index hospitalizations). COPD was defined by ICD-10-CM diagnostic codes, according to Centers for Medicare and Medicaid Services (CMS) readmission measures. The primary outcome was any readmission within 30 days after index hospitalization. In the training set (randomly-selected 70% of sample), patient characteristics and inpatient care data were used as predictors to derive a conventional logistic regression model and two machine learning models (lasso regression and deep neural network). In the test set (remaining 30% of sample), the prediction performances of the machine learning models were examined by comparison with the reference model based on CMS readmission measures. Among 44,929 index hospitalizations for COPD, 3413 (7%) were readmitted within 30 days after discharge. The reference model had the lowest discrimination ability (C-statistic: 0.57 [95% confidence interval (CI) 0.56-0.59]). The two machine learning models had moderate, significantly higher discrimination ability (C-statistic: lasso regression, 0.61 [95% CI 0.59-0.61], p = 0.004; deep neural network, 0.61 [95% CI 0.59-0.63], p = 0.007). Tube feeding duration, blood transfusion, thoracentesis use, and male sex were important predictors. In this study using nationwide administrative data in Japan, machine learning models improved the prediction of 30-day readmission after COPD hospitalization compared with a conventional model.

Treatment costs and cost drivers among osteoporotic fracture patients in Japan: a retrospective database analysis

This study estimated the direct medical costs of osteoporotic fractures from a large claim database in Japan. We further identified several comorbidities which drove the treatment costs. The results would contribute to health economic analysis as well as understanding of individual financial burden in Japan.

INTRODUCTION

The purposes of this study were to estimate treatment costs of osteoporotic fractures and to investigate the cost drivers.

METHODS

Male and female patients aged 50 years and older with a hip, vertebral, or non-hip/non-vert (NHNV) fracture between April 2008 and December 2016 were analyzed from claim database. Two types of costs were estimated. The incremental yearly costs of fractures and comorbidity treatments (total medical costs) were calculated by subtracting pre-fracture costs from post-fracture costs. The costs exclusive for fracture treatments (fracture treatment costs) were estimated by summing up the costs of fracture treatments within 1 year after fracture. The associations between comorbidities and costs were examined with a generalized linear model.

RESULTS

Total 12,898 patients were identified (83% was female). The total medical costs of fractures were $14,592 for male-hip, $15,691 for female-hip, $4268 for male-vertebral, $3819 for female-vertebral, $3790 for male-NHNV, and $4259 for female-NHNV. The fracture treatment costs were $4506 for male-hip, $5427 for female-hip, $1022 for male-vertebral, $1044 for female-vertebral, $1035 for male-NHNV, and $1408 for female NHNV. Three comorbidities were associated with increasing fracture treatment costs whereas four comorbidities were associated with decreasing fracture treatment costs. Five comorbidities were associated with increasing total medical costs whereas one comorbidity was associated with decreasing total medical costs.

CONCLUSIONS

Yearly treatment costs were increased considerably after fracture. Several comorbidities were considered to be cost drivers for osteoporotic fracture treatment. The cost estimates with different patient profile would support conducting health economic analysis in the future.

Impact of hospital volume on mortality in patients with severe torso injury

BACKGROUND

Whether a positive volume-outcome relationship exists in the context of trauma remains controversial. Heterogeneity in the definition of hospital volume in previous studies is one of the main reasons for this inconclusiveness. We investigated whether hospital volume is associated with mortality in patients with severe torso injury using two different definitions of hospital volume.

MATERIALS AND METHODS

This retrospective cohort study used the Diagnosis Procedure Combination database in Japan. Patients who were admitted to tertiary emergency centers with severe torso injury and underwent emergency surgery or interventional radiology treatment for the torso injury upon admission from April 1, 2010 to March 31, 2014 were included. Hospital volume was defined as the annual number of admissions with severe torso injury (HV-torso) or the annual number of total trauma admissions (HV-all). The main outcome was 28-d mortality. Multivariable logistic regression models fitted with generalized estimating equations were used to evaluate relationships between hospital volume and 28-d mortality.

RESULTS

Overall, 7725 patients were included. The 28-d mortality rate was 15.3%. The HV-torso was significantly associated with reduced 28-d mortality (adjusted odds ratio = 0.59; 95% confidence interval = 0.44-0.79). However, there was no significant association between the HV-all and mortality (adjusted odds ratio = 1.02; 95% confidence interval = 0.72-1.46).

CONCLUSIONS

The HV-torso was significantly associated with reduced mortality in patients with severe torso injury. In contrast, the HV-all had no significant relationship with their mortality. Regionalization of trauma care for severe torso injury may be beneficial for patients with severe torso injury.

Associations of Polyethylenimine-Coated AN69ST Membrane in Continuous Renal Replacement Therapy with the Intensive Care Outcomes: Observations from a Claims Database from Japan

BACKGROUND/AIMS

Polyethylenimine-coated polyacrylonitrile (AN69ST) membrane is expected to improve the outcomes of critically ill patients treated by continuous renal replacement therapy (CRRT).

METHODS

Using a Japanese health insurance claim database, we identified adult patients receiving CRRT in intensive care units (ICUs) from April 2014 to October 2015. We used a multivariable logistic regression model to assess in-hospital mortality and Fine and Gray's proportional subhazards model to assess the ICU length of stay (ICU-LOS) accounting for the competing risks.

RESULTS

Of 2,469 ICU patients, 156 were treated by AN69ST membrane. Crude in-hospital mortality was 50.0% in the AN69ST group and 54.0% in the non-AN69ST group. Adjusted odds ratio (OR) of AN69ST membrane use for in-hospital mortality was 0.65 (95% CI 0.45-0.93). The use of AN69ST membrane was also independently associated with shorter ICU-LOS.

CONCLUSION

This retrospective observational study suggested that CRRT with AN69ST membrane might be associated with better in-hospital outcomes.

Outcomes after early and delayed rehabilitation for exacerbation of chronic obstructive pulmonary disease: a nationwide retrospective cohort study in Japan

Background

The effectiveness of early pulmonary rehabilitation (PR) for exacerbation of chronic obstructive pulmonary disease (COPD) remains controversial. The present study aimed to compare the outcomes between early and delayed PR for exacerbation of COPD, using a national inpatient database.

Methods

Using the Japanese Diagnosis Procedure Combination database, we examined patients who were transported to hospital for exacerbation of COPD, received PR during hospitalisation, and were discharged to their home. The patients were divided into those who received early PR (defined as PR starting within 48 h of admission) and those who received delayed PR. The outcomes included 90-day readmission, length of stay (LOS), and activities of daily living (Barthel index ≥15) at discharge. Multiple imputation was used for missing data. To assess the associations between early PR and the outcomes, we used risk-adjusted treatment effects and instrumental variable methods.

Results

We identified 12,572 eligible patients, including 8459 patients with delayed PR and 4113 with early PR. In the risk-adjusted treatment effect models, the early PR group had lower proportion of 90-day readmission (risk difference, −3.4%; 95% CI, −5.7% to −1.5%) and shorter LOS (−9.8 days; 95% CI, −10.8 days to −8.7 days) than the delayed PR group. There was no significant difference in activities of daily living at discharge between the two groups. The instrumental variable analyses showed similar results.

Conclusions

In this national database study, early PR was associated with reduced 90-day readmission and shortened LOS in patients with exacerbation of COPD.

Validity of diagnoses, procedures, and laboratory data in Japanese administrative data

Background

Validation of recorded data is a prerequisite for studies that utilize administrative databases. The present study evaluated the validity of diagnoses and procedure records in the Japanese Diagnosis Procedure Combination (DPC) data, along with laboratory test results in the newly-introduced Standardized Structured Medical Record Information Exchange (SS-MIX) data.

Methods

Between November 2015 and February 2016, we conducted chart reviews of 315 patients hospitalized between April 2014 and March 2015 in four middle-sized acute-care hospitals in Shizuoka, Kochi, Fukuoka, and Saga Prefectures and used them as reference standards. The sensitivity and specificity of DPC data in identifying 16 diseases and 10 common procedures were identified. The accuracy of SS-MIX data for 13 laboratory test results was also examined.

Results

The specificity of diagnoses in the DPC data exceeded 96%, while the sensitivity was below 50% for seven diseases and variable across diseases. When limited to primary diagnoses, the sensitivity and specificity were 78.9% and 93.2%, respectively. The sensitivity of procedure records exceeded 90% for six procedures, and the specificity exceeded 90% for nine procedures. Agreement between the SS-MIX data and the chart reviews was above 95% for all 13 items.

Conclusion

The validity of diagnoses and procedure records in the DPC data and laboratory results in the SS-MIX data was high in general, supporting their use in future studies.

•

Validity of DPC and SS-MIX data was tested using chart review results as references.

•

Specificity of DPC diagnoses was high; sensitivity was low and varied by conditions.

•

Procedure records in DPC data were highly sensitive and specific.

•

Accuracy of SS-MIX laboratory data exceeded 95% in all 13 items examined.