Validity of Using Japanese Administrative Data to Identify Inpatients With Acute Pulmonary Embolism: Referencing the COMMAND VTE Registry

BACKGROUND

Acute pulmonary embolism (PE) is a life-threatening in-hospital complication. Recently, several studies have reported the clinical characteristics of PE among Japanese patients using the diagnostic procedure combination (DPC)/per diem payment system database. However, the validity of PE identification algorithms for Japanese administrative data is not yet clear. The purpose of this study was to evaluate the validity of using DPC data to identify acute PE inpatients.

METHODS

The reference standard was symptomatic/asymptomatic PE patients included in the COntemporary ManageMent AND outcomes in patients with Venous ThromboEmbolism (COMMAND VTE) registry, which is a cohort study of acute symptomatic venous thromboembolism (VTE) patients in Japan. The validation cohort included all patients discharged from the six hospitals included in both the registry and DPC database. The identification algorithms comprised diagnosis, anticoagulation therapy, thrombolysis therapy, and inferior vena cava filter placement. Each algorithm's sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were estimated.

RESULTS

A total of 43.4% of the validation cohort was female, with a mean age of 67.3 years. The diagnosis-based algorithm showed a sensitivity of 90.2% (222/246; 95% confidence interval [CI], 85.8-93.6%), a specificity of 99.8% (228,485/229,027; 95% CI, 99.7-99.8%), a PPV of 29.1% (222/764; 95% CI, 25.9-32.4%) and an NPV of 99.9% (228,485/229,509; 95% CI, 99.9-99.9%) for identifying symptomatic/asymptomatic PE. Additionally, 94.6% (159/168; 95% CI, 90.1-97.5%) of symptomatic PE patients were identified using the diagnosis-based algorithm.

CONCLUSION

The diagnosis-based algorithm may be a relatively sensitive method for identifying acute PE inpatients in the Japanese DPC database.

Quantified trends in provider reimbursement and utilization volume in colorectal cancer screenings: analysis of Medicare claims from 2000 to 2019

AIM

Colorectal cancer (CRC) screening rates in the United States remain persistently below guideline targets, partly due to suboptimal patient utilization and provider reimbursement. To guide long-term national utilization estimates and set reasonable screening adherence targets, this study aimed to quantify trends in utilization of and reimbursement for CRC screenings using Medicare claims.

METHOD

Inflation-adjusted reimbursements and utilization volume associated with each CRC screening code were abstracted from Medicare claims between 2000 and 2019. Screenings, screenings/100 000 enrolees and reimbursement/screening were analysed with linear regression and compared with the equality of slopes tests. Average reimbursement per screening was compared using analysis of variance with Dunnett's T3 multiple comparisons test.

RESULTS

The growth rate of multitarget stool DNA tests (mt-sDNA)/100 000 was the highest at 170.4 screenings/year (R2 = 0.99, p ≤ 0.001), while that of faecal occult blood tests/100 000 was the lowest at -446.4 screenings/year (R2 = 0.90, p ≤ 0.001) (p ≤ 0.001). Provider reimbursements averaged $546.95 (95% CI $520.12-$573.78) per mt-sDNA screening, significantly higher than reimbursements for all invasive screenings. Only FOBTs significantly increased in reimbursement per screening at $0.62/year (R2 = 0.91, p ≤ 0.001).

CONCLUSION

We derived forecastable trend numbers for utilization and provider reimbursement. Faecal immunochemical tests/100 000 and mt-sDNA screenings/100 000 increased most rapidly during the entire study period. The number of nearly all invasive screenings/100 000 decreased rapidly; the number of colonoscopies/100 000 increased slightly, probably due to superior diagnostic strength. These trends indicate the that replacement of other invasive modalities with accessible noninvasive screenings will account for much of future screening behaviour and thus reductions in CRC incidence and mortality, especially given providers' reimbursement incentive to screen average-risk patients with stool-based tests.

A panel vector autoregression analysis for the dynamics of medical and long-term care expenditures

Although medical and long-term care expenditures for older adults are closely related, providing rigorous statistical analysis for their dynamic relationship is challenging. In this research, we propose a novel approach using the panel vector autoregression model to reveal the realized patterns of the interdependence. As an empirical application, we analyze monthly panel data on individuals in a city of Japan, where social insurance covers many formal services for long-term care. Our estimation results indicate the existence of intertemporal transition from expensive acute medical care to reasonable at-home medical care, then to at-home long-term care. Under this context, the enhancement of formal long-term care sector in Japan might have played an important role in the suppression of the total care cost in spite for its rapid aging over the past 2 decades. Additionally, we find that daycare plays multiple roles in Japanese long-term care, such as respite and rehabilitation, but there is no considerable transition from outpatient rehabilitation to daycare in the long-term care sector.

Factors associated with the introduction of visiting-pharmacist services in older adults in Japan: A nested case-control study

AIM

To investigate the factors associated with introducing visiting-pharmacist services for community-dwelling older adults in Japan.

METHODS

We conducted a nested case-control study using claims data in a cohort from a city in Tokyo. Patients aged ≥65 years who received visiting-pharmacist services for the first time between April 2014 and March 2020 were considered case patients. A maximum of four controls to each case patient were randomly selected on the basis of sex, age, health insurance systems, and month-year. Medical and long-term care service usage and patient condition were assessed using claims data from the index and preceding months, along with long-term care needs certification data. Multivariable conditional logistic regression analysis was conducted to estimate the adjusted odds ratios with 95% confidence intervals for factors associated with visiting-pharmacist service introduction.

RESULTS

A total of 22 949 participants (4591 cases and 18 358 controls) were included, with a median age of 85 years; 59.3% were women. The adjusted odds ratios (95% confidence intervals) of the three most related factors were 27.61 (23.98-31.80) for physicians' home visits, 5.83 (5.08-6.70) for hospitalization, and 4.97 (4.16-5.95) for designated-facility admission. Factors such as prescribing ≧10 medications, visiting nursing, and cancer were positively associated. In contrast, low household income and a high need for support due to cognitive function or disability were negatively associated.

CONCLUSIONS

This study provides insights into the introduction of visiting-pharmacist services for older adults in Japan. Geriatr Gerontol Int 2024; 24: 344-351.

Drug-drug interaction potential among patients with nonmetastatic castration-resistant prostate cancer (nmCRPC) treated with novel androgen receptor inhibitors

BACKGROUND

Nonmetastatic castration-resistant prostate cancer (nmCRPC) patients are often older and use concurrent medications that increase the potential for drug-drug interactions (pDDIs). This study assessed pDDI prevalence in real-world nmCRPC patients treated with apalutamide, darolutamide, or enzalutamide.

RESEARCH DESIGN AND METHODS

Castrated prostate cancer patients without metastases prior to androgen receptor inhibitor initiation were identified retrospectively via Optum Clinformatics Data Mart claims data (8/2019-3/2021). The top 100 concomitant medications were assessed for pDDIs.

RESULTS

Among 1,515 patients (mean age: 77 ± 8 years; mean Charlson Comorbidity Index: 3 ± 3), 340 initiated apalutamide, 112 darolutamide, and 1,063 enzalutamide. Common concomitant medication classes were cardiovascular (80%) and central nervous system (52%). Two-thirds of the patients received ≥5 concomitant medications; 30 (30/100 medications) pDDIs were identified for apalutamide and enzalutamide each and 2 (2/100 medications) for darolutamide. Most pDDIs had risk ratings of C or D, but four for apalutamide were rated X. Approximately 58% of the patients on apalutamide, 5% on darolutamide, and 54% on enzalutamide had ≥1 identified pDDI.

CONCLUSIONS

Results showed a higher frequency of pDDIs in patients receiving apalutamide and enzalutamide vs darolutamide. The impact of these could not be determined retrospectively. DDI risk should be carefully evaluated when discussing optimal therapy for patients with nmCRPC.

Predicting pediatric severe asthma exacerbations: an administrative claims-based predictive model

OBJECTIVE

Previous machine learning approaches fail to consider race and ethnicity and social determinants of health (SDOH) to predict childhood asthma exacerbations. A predictive model for asthma exacerbations in children is developed to explore the importance of race and ethnicity, rural-urban commuting area (RUCA) codes, the Child Opportunity Index (COI), and other ICD-10 SDOH in predicting asthma outcomes.

METHODS

Insurance and coverage claims data from the Arkansas All-Payer Claims Database were used to capture risk factors. We identified a cohort of 22,631 children with asthma aged 5-18 years with 2 years of continuous Medicaid enrollment and at least one asthma diagnosis in 2018. The goal was to predict asthma-related hospitalizations and asthma-related emergency department (ED) visits in 2019. The analytic sample was 59% age 5-11 years, 39% White, 33% Black, and 6% Hispanic. Conditional random forest models were used to train the model.

RESULTS

The model yielded an area under the curve (AUC) of 72%, sensitivity of 55% and specificity of 78% in the OOB samples and AUC of 73%, sensitivity of 58% and specificity of 77% in the training samples. Consistent with previous literature, asthma-related hospitalization or ED visits in the previous year (2018) were the two most important variables in predicting hospital or ED use in the following year (2019), followed by the total number of reliever and controller medications.

CONCLUSIONS

Predictive models for asthma-related exacerbation achieved moderate accuracy, but race and ethnicity, ICD-10 SDOH, RUCA codes, and COI measures were not important in improving model accuracy.

A Novel Method for Assessing Risk-Adjusted Diagnostic Coding Specificity for Depression Using a U.S. Cohort of over One Million Patients

Depression is a prevalent and debilitating mental health condition that poses significant challenges for healthcare providers, researchers, and policymakers. The diagnostic coding specificity of depression is crucial for improving patient care, resource allocation, and health outcomes. We propose a novel approach to assess risk-adjusted coding specificity for individuals diagnosed with depression using a vast cohort of over one million inpatient hospitalizations in the United States. Considering various clinical, demographic, and socioeconomic characteristics, we develop a risk-adjusted model that assesses diagnostic coding specificity. Results demonstrate that risk-adjustment is necessary and useful to explain variability in the coding specificity of principal (AUC = 0.76) and secondary (AUC = 0.69) diagnoses. Our approach combines a multivariate logistic regression at the patient hospitalization level to extract risk-adjusted probabilities of specificity with a Poisson Binomial approach at the facility level. This method can be used to identify healthcare facilities that over- and under-specify diagnostic coding when compared to peer-defined standards of practice.

Incidence and Prevalence of Epilepsy in Japan: A Retrospective Analysis of Insurance Claims Data of 9,864,278 Insured Persons

INTRODUCTION

The burden of epilepsy is thought to be high but is difficult to measure. Very few studies in Japan have attempted to estimate prevalence and incidence rates of epilepsy in Japan.

METHODS

This retrospective cohort study used commercially collected nationwide insurance claims data from a cohort of 10 million persons between 2012 and 2019 among those aged 0 to 74 years. Using the claims data, cases were identified, and incidence and prevalence rates were estimated.

RESULTS

A total of 9,864,278 persons were included. The average age was 34.5 (standard deviation, 18.5) years. A total of 77,312 persons were diagnosed with epilepsy over the 8-year observation period, with a prevalence rate of 6.0 per 1,000 persons with almost no difference by gender. The highest rates were seen among those aged 70-74 years; prevalence rates tended to rise with calendar year (5.4/1,000 in 2012 and 6.0/1,000 in 2019). The incidence rate of epilepsy was 72.1 per 100,000 person-years with slightly higher rates seen among females. Incidence rates were highest at ages less than 12 months (199.8/100,000 person-years), followed by the eldest age group (70-74 years, 179.4/100,000 person-years).

CONCLUSION

Understanding the magnitude of disease burden is the basis of determining health policies. In this study, the prevalence and incidence of epilepsy in Japan was shown based on the analysis results of a large-scale general population insurance claims data covering all over Japan.

Development of Cross-Sector Health Care Quality Indicators Based on Claims Data

Quality indicators serve as a tool to measure and improve evidence-based health care. Often the transition from inpatient to outpatient care is not sufficiently included. The focus of this work was to develop and to evaluate methods to define relevant cross-sector quality indicators based on Austrian claims data, using myocardial infarction as tracer.

Disease Trends in Children and Adolescents in Japan: A Retrospective Observational Study Using the Nationwide Claims Data for 2012-2016

This study aimed to clarify diseases that occur more frequently by age and identify the peaks and trends of each disease from infancy to adolescence for early detection and treatment. This retrospective observational study was conducted using Japan's National Database of Health Insurance Claims Specific Health Checkups from January 2012 to December 2016. Using peak ages and trends in the number of patients, we grouped diseases by the International Classification of Diseases chapters. Although diseases that peaked during infancy were the most common (10 disease chapters), other diseases peaked at school-going age and adolescence. Diseases in four chapters peaked during adolescence and continued to increase toward the age of 18. These four chapters included mental, behavioral, and neurodevelopmental disorders; diseases of the nervous system; the genitourinary system; and pregnancy, childbirth, and the puerperium. Childhood-onset diseases can affect long-term health and healthcare needs, and timely screening and guidance based on disease trends can provide an effective intervention. To establish a child healthcare system that provides preventive support for children and adolescents' physical, psychological, and social health, further research is needed to comprehensively understand the issues per age and developmental stage.

Pediatric and Young Adult Household Transmission of the Initial Waves of SARS-CoV-2 in the United States: Administrative Claims Study

BACKGROUND

The correlates responsible for the temporal changes of intrahousehold SARS-CoV-2 transmission in the United States have been understudied mainly due to a lack of available surveillance data. Specifically, early analyses of SARS-CoV-2 household secondary attack rates (SARs) were small in sample size and conducted cross-sectionally at single time points. From these limited data, it has been difficult to assess the role that different risk factors have had on intrahousehold disease transmission in different stages of the ongoing COVID-19 pandemic, particularly in children and youth.

OBJECTIVE

This study aimed to estimate the transmission dynamic and infectivity of SARS-CoV-2 among pediatric and young adult index cases (age 0 to 25 years) in the United States through the initial waves of the pandemic.

METHODS

Using administrative claims, we analyzed 19 million SARS-CoV-2 test records between January 2020 and February 2021. We identified 36,241 households with pediatric index cases and calculated household SARs utilizing complete case information. Using a retrospective cohort design, we estimated the household SARS-CoV-2 transmission between 4 index age groups (0 to 4 years, 5 to 11 years, 12 to 17 years, and 18 to 25 years) while adjusting for sex, family size, quarter of first SARS-CoV-2 positive record, and residential regions of the index cases.

RESULTS

After filtering all household records for greater than one member in a household and missing information, only 36,241 (0.85%) of 4,270,130 households with a pediatric case remained in the analysis. Index cases aged between 0 and 17 years were a minority of the total index cases (n=11,484, 11%). The overall SAR of SARS-CoV-2 was 23.04% (95% CI 21.88-24.19). As a comparison, the SAR for all ages (0 to 65+ years) was 32.4% (95% CI 32.1-32.8), higher than the SAR for the population between 0 and 25 years of age. The highest SAR of 38.3% was observed in April 2020 (95% CI 31.6-45), while the lowest SAR of 15.6% was observed in September 2020 (95% CI 13.9-17.3). It consistently decreased from 32% to 21.1% as the age of index groups increased. In a multiple logistic regression analysis, we found that the youngest pediatric age group (0 to 4 years) had 1.69 times (95% CI 1.42-2.00) the odds of SARS-CoV-2 transmission to any family members when compared with the oldest group (18 to 25 years). Family size was significantly associated with household viral transmission (odds ratio 2.66, 95% CI 2.58-2.74).

CONCLUSIONS

Using retrospective claims data, the pediatric index transmission of SARS-CoV-2 during the initial waves of the COVID-19 pandemic in the United States was associated with location and family characteristics. Pediatric SAR (0 to 25 years) was less than the SAR for all age other groups. Less than 1% (n=36,241) of all household data were retained in the retrospective study for complete case analysis, perhaps biasing our findings. We have provided measures of baseline household pediatric transmission for tracking and comparing the infectivity of later SARS-CoV-2 variants.

Prevalence of psoriasis and psoriatic arthritis in Germany - analysis of claims data

BACKGROUND AND OBJECTIVES

Few studies are available on the epidemiology of psoriasis and psoriatic arthritis in Germany. The aim of this study was to estimate this prevalence based on different case definitions.

METHODS

Statutory health insurance (SHI) data were examined (2017 to 2019). Prevalence was analyzed and validated using three different case definitions: (1) At least one inpatient or outpatient diagnosis within one year, (2) additionally at least two outpatient diagnoses within one year, (3) additionally within three years.

RESULTS

The administrative prevalence of psoriasis including psoriatic arthritis ranged from 1.90% to 2.51%. For all case definitions, the prevalence increased with age, decreasing from the age of 70 with psoriasis and from the age of 65 with psoriatic arthritis. Males were more likely to be affected at an older age (p <0.0001), while in the under-20 age group, more girls were affected (p  =  0.04).

CONCLUSIONS

Psoriasis is a common skin disease in Germany. The internal diagnoses validation showed that in future studies with claims data, the narrow and broad criteria should be used to identify patients with psoriasis, depending on the research question.

Study of the Direct Costs of Asthma at a Healthcare Service Provider in Bogotá, Colombia

OBJECTIVES

This study aimed to describe clinical characteristics and direct medical costs associated with disease treatment in Colombia patients with asthma from 1 healthcare provider.

METHODS

This was a descriptive study with a retrospective data collection from a healthcare provider's electronic medical records in Colombia. A clinical, demographic, and healthcare resource utilization profile was developed over a 12-month observation period after the identification of eligible patients. To determine the mean cost per patient per year, the total frequencies of resource utilization were added, and the result was multiplied by the unit cost of each of them.

RESULTS

A total of 7919 patients were included in the analysis. The mean ± SD cost per patient per year ranged from $189.5 ± $1.900.6 to $240.2 ± $1.903.6 depending on the price guidebook. The total cost had been driven by the medication use (79% of total cost) and by the outpatient visits (20% of total cost).

CONCLUSIONS

In the population analyzed, the mean total direct cost per patient per year of asthma was $189.5 and $240.2, depending on the cost source. Direct medical costs were higher in cases classified as severe and in the adult and elderly population. When comparing the sources of resource utilization, it was found that the mean cost per patient obtained from real-life data is lower than the theoretical cost obtained from the bottom-up method with quantification of resources from experts. It is important to consider limitations related to study design and the evolving landscape of asthma treatments.

Use of clinical classifications software to address ICD coding transition in large healthcare databases analyzed via high-dimensional propensity scores

PURPOSE

The EUPAS26595 study characterized the rate of acute renal failure (ARF) in patients exposed to levetiracetam versus other antiepileptic drugs using healthcare claims data and a high-dimensional propensity score (hd-PS) for confounding adjustment. The data contained several coding systems by design and an update in International Classification of Diseases (ICD) coding dictionary. Such coding heterogeneity can affect the performance of hd-PS, and manually coding harmonization is not feasible. Our objective was to explore the impact of code aggregation via Clinical Classifications Software (CCS) on the analysis of a large claims-based database using hd-PS.

METHODS

Patients with epilepsy, who were new-users of an antiepileptic drug, were identified from the IBM® MarketScan® Research Databases. We used CCS categories to harmonize coding and compared the results with other alternatives. Incidence rate ratios (IRRs) were computed using modified Poisson regression model with a robust variance estimator.

RESULTS

For January 2008-October 2015 (before ICD update), 34 833 eligible patients initiated levetiracetam and 52 649 initiated a comparator drug; IRR (95% CI) for ARF for the hd-PS analysis was 1.34 (0.72-2.50) without CCS categories and 1.30 (0.71-2.39) with CCS categories. For January 2008-December 2017 (including ICD coding change), 45 672 eligible patients initiated levetiracetam and 64 664 initiated a comparator drug; IRR (95% CI) for the hd-PS analysis was 1.34 (0.78-2.29) without CCS categories and 1.37 (0.80-2.34) with CCS categories.

CONCLUSIONS

Using single-level CCS categories to overcome differences in coding provides consistent results and can be used in studies that use large claims data and hd-PS for adjustment.

Healthcare resource utilization and associated costs in patients with metastatic urothelial carcinoma: a real-world analysis using German claims data

AIMS

This retrospective claims data study characterized real-world treatment patterns, healthcare resource utilization (HCRU), and costs in patients with metastatic urothelial carcinoma (mUC) in Germany.

MATERIALS AND METHODS

Continuously insured adults with incident mUC diagnosis (=index; ICD-10: C65-C68/C77-C79) in 2015-2019 were identified from two German claims databases. Patients who received first-line (1 L) treatment within 12 months of index were divided into three mutually exclusive sub-cohorts: platinum-based chemotherapy (PB-CT), non-PB-CT, and immunotherapy (IO). Patient characteristics were assessed during a 24-month baseline period; treatments, HCRU, and costs (of the health insurance fund) per patient-year (ppy) were described during 12-month follow-up.

RESULTS

We identified 3,226 patients with mUC (mean age, 73.8 years; male, 70.8%; mean Elixhauser Comorbidity Index, 17.6); 1,286 (39.9%) received 1 L treatment within 12 months of index. Of these, 825 (64.2%) received PB-CT, 322 (25.0%) non-PB-CT, and 139 (10.8%) IO. On average, treated patients had 5.1 hospitalizations ppy. Most UC-related hospitalizations ppy were observed in the PB-CT cohort (5.8), followed by the non-PB-CT (4.2) and IO (2.3) cohorts. Mean UC-related hospitalization costs ppy were €22,218 in the treated cohort, €24,294 in PB-CT, €19,079 in IO, and €18,530 in non-PB-CT cohorts. Cancer-related prescription costs ppy averaged €6,323 in treated patients, and €25,955 in IO, €4,318 in non-PB-CT, and €4,270 in PB-CT cohorts.

LIMITATIONS

We recognized limitations in our study's sample selection due to unavailable mUC disease status data. We addressed this through an upstream feasibility study conducted in consultation with clinical experts to determine a suitable proxy. Proxies were also used to delineate treatment lines, switches, and discontinuations due to data absence. Furthermore, due to data restrictions, collective dataset analysis was not possible, prompting a meta-analysis for pooled results.

CONCLUSIONS

The study shows that mUC is associated with significant HCRU and costs across different types of 1 L systemic therapy.

Prävalenz der Psoriasis und Psoriasis-Arthritis in Deutschland - Analyse von Routinedaten der gesetzlichen Krankenversicherung: Prevalence of psoriasis and psoriatic arthritis in Germany - analysis of claims data

Hintergrund und Zielsetzung

Es gibt nur wenige Studien zur Epidemiologie der Psoriasis und Psoriasis‐Arthritis in Deutschland. Ziel dieser Studie war es, diese Prävalenz anhand verschiedener Falldefinitionen zu schätzen.

Methoden

Es wurden Daten der Gesetzlichen Krankenversicherung (GKV) untersucht (2017 bis 2019). Die Prävalenz wurde anhand von drei verschiedenen Falldefinitionen analysiert und validiert: (1) mindestens eine stationäre oder ambulante Diagnose innerhalb eines Jahres (2) mindestens zwei ambulante Diagnosen innerhalb eines Jahres (3) zusätzlich innerhalb von drei Jahren.

Ergebnisse

Die administrative Prävalenz der Psoriasis inklusive Psoriasis‐Arthritis lag zwischen 1,90% und 2,51%. Bei allen Falldefinitionen nahm die Prävalenz mit dem Alter zu, wobei sie bei der Psoriasis ab einem Alter von 70 Jahren und bei der Psoriasis‐Arthritis ab einem Alter von 65 Jahren sank. Männer waren im höheren Alter häufiger betroffen (p <0,0001), während in der Altersgruppe der unter 20‐Jährigen mehr Mädchen betroffen waren (p  =  0,04).

Schlussfolgerung

Die Psoriasis ist eine häufige Hauterkrankung in Deutschland. Die interne Diagnosenvalidierung hat gezeigt, dass in zukünftigen Studien mit GKV‐Routinedaten, je nach Fragestellung, die engen oder weiten Kriterien zur Identifizierung von Patienten mit Psoriasis verwendet werden sollten.

Drug prescription patterns and their association with mortality and hospitalization duration in COVID-19 patients: insights from big data

Background

Different medication prescription patterns have been associated with varying course of disease and outcomes in COVID-19. Health claims data is a rich source of information on disease treatment and outcomes. We aimed to investigate drug prescription patterns and their association with mortality and hospitalization via insurance data for a relatively long period of the pandemic in Iran.

Methods

We retrieved hospitalized patients' data from Iran Health Insurance Organization (IHIO) spanning 26 months (2020-2022) nationwide. Included were patients with ICD-10 codes U07.1/U07.2 for confirmed/suspected COVID-19. A case was defined as a single hospitalization event for an individual patient. Multiple hospitalizations of a patient within a 30-day interval were aggregated into a single case, while hospitalizations with intervals exceeding 30 days were treated as independent cases. The Anatomical Therapeutic Chemical (ATC) was used for medications classification. The two main study outcomes were general and intensive care unit (ICU) hospitalization periods and mortality. Besides, various demographic and clinical associate factors were analyzed to derive the associations with medication prescription patterns and study outcomes using accelerated failure time (AFT) and logistic regression models.

Results

During the 26 months of the study period, 1,113,678 admissions with COVID-19 diagnosis at hospitals working in company with IHIO were recorded. 917,198 cases were detected from the database, among which 51.91% were females and 48.09% were males. Among the main groups of medications, antithrombotics (55.84% [95% CI: 55.74-55.94]), corticosteroids (54.14% [54.04-54.24]), and antibiotics (42.22% [42.12-42.32]) were the top used medications among cases with COVID-19. Investigation of the duration of hospitalization based on main medication groups showed antithrombotics (adjusted median ratio = 0.94 [0.94-0.95]) were significantly associated with shorter periods of overall hospitalization. Also, antithrombotics (adjusted odds ratio = 0.74 [95%CI, 0.73-0.76]), corticosteroids (0.97 [0.95-0.99]), antivirals (0.82 [0.80-0.83]), and ACE inhibitor/ARB (0.79 [0.77-0.80]) were significantly associated with lower mortality.

Conclusion

Over 2 years of investigation, antithrombotics, corticosteroids, and antibiotics were the top medications for hospitalized patients with COVID-19. Trends in medication prescription varied based on various factors across the country. Medication prescriptions could potentially significantly impact the trends of mortality and hospitalization during epidemics, thereby affecting both health and economic burdens.

Avenues for Strengthening PCORnet's Capacity to Advance Patient-Centered Economic Outcomes in Patient-Centered Outcomes Research (PCOR)

PCORnet, the National Patient-Centered Clinical Research Network, provides the ability to conduct prospective and observational pragmatic research by leveraging standardized, curated electronic health records data together with patient and stakeholder engagement. PCORnet is funded by the Patient-Centered Outcomes Research Institute (PCORI) and is composed of 8 Clinical Research Networks that incorporate at total of 79 health system "sites." As the network developed, linkage to commercial health plans, federal insurance claims, disease registries, and other data resources demonstrated the value in extending the networks infrastructure to provide a more complete representation of patient's health and lived experiences. Initially, PCORnet studies avoided direct economic comparative effectiveness as a topic. However, PCORI's authorizing law was amended in 2019 to allow studies to incorporate patient-centered economic outcomes in primary research aims. With PCORI's expanded scope and PCORnet's phase 3 beginning in January 2022, there are opportunities to strengthen the network's ability to support economic patient-centered outcomes research. This commentary will discuss approaches that have been incorporated to date by the network and point to opportunities for the network to incorporate economic variables for analysis, informed by patient and stakeholder perspectives. Topics addressed include: (1) data linkage infrastructure; (2) commercial health plan partnerships; (3) Medicare and Medicaid linkage; (4) health system billing-based benchmarking; (5) area-level measures; (6) individual-level measures; (7) pharmacy benefits and retail pharmacy data; and (8) the importance of transparency and engagement while addressing the biases inherent in linking real-world data sources.

Identifying Predictors of Nursing Home Admission by Using Electronic Health Records and Administrative Data: Scoping Review

Background

Among older adults, nursing home admissions (NHAs) are considered a significant adverse outcome and have been extensively studied. Although the volume and significance of electronic data sources are expanding, it is unclear what predictors of NHA have been systematically identified in the literature via electronic health records (EHRs) and administrative data.

Objective

This study synthesizes findings of recent literature on identifying predictors of NHA that are collected from administrative data or EHRs.

Methods

The PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines were used for study selection. The PubMed and CINAHL databases were used to retrieve the studies. Articles published between January 1, 2012, and March 31, 2023, were included.

Results

A total of 34 papers were selected for final inclusion in this review. In addition to NHA, all-cause mortality, hospitalization, and rehospitalization were frequently used as outcome measures. The most frequently used models for predicting NHAs were Cox proportional hazards models (studies: n=12, 35%), logistic regression models (studies: n=9, 26%), and a combination of both (studies: n=6, 18%). Several predictors were used in the NHA prediction models, which were further categorized into sociodemographic, caregiver support, health status, health use, and social service use factors. Only 5 (15%) studies used a validated frailty measure in their NHA prediction models.

Conclusions

NHA prediction tools based on EHRs or administrative data may assist clinicians, patients, and policy makers in making informed decisions and allocating public health resources. More research is needed to assess the value of various predictors and data sources in predicting NHAs and validating NHA prediction models externally.

Assessing the Use of German Claims Data Vocabularies for Research in the Observational Medical Outcomes Partnership Common Data Model: Development and Evaluation Study

Background

National classifications and terminologies already routinely used for documentation within patient care settings enable the unambiguous representation of clinical information. However, the diversity of different vocabularies across health care institutions and countries is a barrier to achieving semantic interoperability and exchanging data across sites. The Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM) enables the standardization of structure and medical terminology. It allows the mapping of national vocabularies into so-called standard concepts, representing normative expressions for international analyses and research. Within our project "Hybrid Quality Indicators Using Machine Learning Methods" (Hybrid-QI), we aim to harmonize source codes used in German claims data vocabularies that are currently unavailable in the OMOP CDM.

Objective

This study aims to increase the coverage of German vocabularies in the OMOP CDM. We aim to completely transform the source codes used in German claims data into the OMOP CDM without data loss and make German claims data usable for OMOP CDM-based research.

Methods

To prepare the missing German vocabularies for the OMOP CDM, we defined a vocabulary preparation approach consisting of the identification of all codes of the corresponding vocabularies, their assembly into machine-readable tables, and the translation of German designations into English. Furthermore, we used 2 proposed approaches for OMOP-compliant vocabulary preparation: the mapping to standard concepts using the Observational Health Data Sciences and Informatics (OHDSI) tool Usagi and the preparation of new 2-billion concepts (ie, concept_id >2 billion). Finally, we evaluated the prepared vocabularies regarding completeness and correctness using synthetic German claims data and calculated the coverage of German claims data vocabularies in the OMOP CDM.

Results

Our vocabulary preparation approach was able to map 3 missing German vocabularies to standard concepts and prepare 8 vocabularies as new 2-billion concepts. The completeness evaluation showed that the prepared vocabularies cover 44.3% (3288/7417) of the source codes contained in German claims data. The correctness evaluation revealed that the specified validity periods in the OMOP CDM are compliant for the majority (705,531/706,032, 99.9%) of source codes and associated dates in German claims data. The calculation of the vocabulary coverage showed a noticeable decrease of missing vocabularies from 55% (11/20) to 10% (2/20) due to our preparation approach.

Conclusions

By preparing 10 vocabularies, we showed that our approach is applicable to any type of vocabulary used in a source data set. The prepared vocabularies are currently limited to German vocabularies, which can only be used in national OMOP CDM research projects, because the mapping of new 2-billion concepts to standard concepts is missing. To participate in international OHDSI network studies with German claims data, future work is required to map the prepared 2-billion concepts to standard concepts.

Identifying inpatient mortality in MarketScan claims data using machine learning

PURPOSE

Inpatient mortality is an important variable in epidemiology studies using claims data. In 2016, MarketScan data began obscuring specific hospital discharge status types for patient privacy, including inpatient deaths, by setting the values to missing. We used a machine learning approach to correctly identify hospitalizations that resulted in inpatient death using data prior to 2016.

METHODS

All hospitalizations from 2011 to 2015 with discharge status of missing, died, or one of the other subsequently obscured values were identified and divided into a training set and two test sets. Predictor variables included age, sex, elapsed time from hospital discharge until last observed claim and until healthcare plan disenrollment, and absence of any discharge diagnoses. Four machine learning methods were used to train statistical models and assess sensitivity and positive predictive value (PPV) for inpatient mortality.

RESULTS

Overall 1 307 917 hospitalizations were included. All four machine learning approaches performed well in all datasets. Random forest performed best with 88% PPV and 93% sensitivity for the training set and both test sets. The two factors with the highest relative importance for identifying inpatient mortality were having no observed claims for the patient on days 2-91 following hospital discharge and patient disenrollment from the healthcare plan within 60 days following hospital discharge.

CONCLUSION

We successfully developed machine learning algorithms to identify inpatient mortality. This approach can be applied to obscured data to accurately identify inpatient mortality among hospitalizations with missing discharge status.

Impact of COVID-19 measures on exacerbation rates and healthcare visits in US asthma patients

Objective: To compare exacerbation rates and healthcare resource utilization (HCRU) in real-world patients in the United States who had moderate-to-severe asthma on medium- or high-dose inhaled corticosteroid/long-acting β₂-agonist therapy at different stages before and after the pandemic. Methods: This noninterventional, retrospective study described demographics, exacerbations, HCRU, and medication use in patients from a US-wide healthcare claims database in 4 consecutive years anchored around March 15, 2020 (start date of the first emergency health measures against coronavirus disease 2019 [COVID-19], or the first lockdown, in the United States, termed "restriction onset" hereafter). Four cohorts of patients potentially eligible for moderate-to-severe asthma clinical trials at the beginning (index) of each of four 1-year periods (March 15, 2018, 2019, 2020, 2021, respectively) were built. Exacerbations, healthcare visits, and asthma medication use were counted in the 1-year period after the index for each cohort. Results: The prevalence of patients with one or more exacerbation per year decreased by 10.00% in the first year after the restriction onset compared with the year before and attenuated over time to 6.37% in the second year. The proportion of inpatient, emergency department, and physician's office visits remained stable over the time periods evaluated for all patients and those patients who experienced one or more exacerbations. Asthma treatment of patients who experienced one or more exacerbations also remained stable over the 4 years. Conclusion: The effect of COVID-19 public health measures on asthma exacerbation rates might have affected clinical trials being run during this period and should be considered in their analysis. Asthma clinical trials run under pandemic hygiene restrictions should consider lower exacerbation frequency in their study design, while treatment and healthcare visits seem unchanged.

Scoping Review of Employer-Led Research Using Employee Health Claims Data

Employers may evaluate employee claims data for various reasons, including assessment of medical insurance and wellness plan efficacy, monitoring employee health trends, and identifying focus areas for wellness measures. The objective of this scoping review (ScR) is to describe the available literature reporting the use, applications, and outcomes of employee health claims data by self-insured employers. The ScR was conducted in a stepwise manner using an established framework: identifying the research question, identifying and selecting relevant studies, charting the data, and collating and reporting results. Literature searches were conducted in PubMed and Embase. Studies of self-insured employee populations that were conducted by the employer/s through May 2022 were identified using predefined criteria. Forty-one studies were included. The majority (90%) were cohort study designs; most employers (51%) were in industries such as aluminum production and health insurance providers. Twenty-four (59%) studies supplemented claims data with other sources such as human resource data to evaluate programs and/or health outcomes. A range of exposures (eg, chronic conditions, wellness program participation) and outcomes (eg, rates or costs of conditions, program effectiveness) were considered. Among the 25 studies that reported on patient confidentiality and privacy, 68% indicated institutional review board approval and 48% reported use of deidentified data. Many self-insured employers have used employee health claims data to gain insights into their employees' needs and health care utilization. These data can be used to identify potential improvements for wellness and other targeted programs to improve employee health and decrease absenteeism.

Postacute Expenditures Among Patients Discharged Home After Stroke or Transient Ischemic Attack: The COMprehensive Post-Acute Stroke Services (COMPASS) Trial

OBJECTIVES

The COMPASS (COMprehensive Post-Acute Stroke Services) pragmatic trial cluster-randomized 40 hospitals in North Carolina to the COMPASS transitional care (TC) postacute care intervention or usual care. We estimated the difference in healthcare expenditures postdischarge for patients enrolled in the COMPASS-TC model of care compared with usual care.

METHODS

We linked data for patients with stroke or transient ischemic attack enrolled in the COMPASS trial with administrative claims from Medicare fee-for-service (n = 2262), Medicaid (n = 341), and a large private insurer (n = 234). The primary outcome was 90-day total expenditures, analyzed separately by payer. Secondary outcomes were total expenditures 30- and 365-days postdischarge and, among Medicare beneficiaries, expenditures by point of service. In addition to intent-to-treat analysis, we conducted a per-protocol analysis to compare Medicare patients who received the intervention with those who did not, using randomization status as an instrumental variable.

RESULTS

We found no statistically significant difference in total 90-day postacute expenditures between intervention and usual care; the results were consistent across payers. Medicare beneficiaries enrolled in the COMPASS intervention arm had higher 90-day hospital readmission expenditures ($682, 95% CI $60-$1305), 30-day emergency department expenditures ($132, 95% CI $13-$252), and 30-day ambulatory care expenditures ($67, 95% CI $38-$96) compared with usual care. The per-protocol analysis did not yield a significant difference in 90-day postacute care expenditures for Medicare COMPASS patients.

CONCLUSIONS

The COMPASS-TC model did not significantly change patients' total healthcare expenditures for up to 1 year postdischarge.

Influenza vaccination uptake among at-risk patients in Switzerland-The potential of national claims data for surveillance

Background

Swiss national surveillance of influenza vaccination uptake rates (VURs) relies on self-reported vaccination status. The aim of this study was to determine VURs among at-risk patients, namely, patients ≥65 of age and adult patients with chronic diseases, using claims data, instead of self-reported measures, to investigate factors of vaccine uptake, and to assess different methodological approaches to conduct vaccination surveillance.

Methods

In this retrospective cross-sectional analysis, we determined VURs in three influenza seasons (2015/2016-2017/2018). Medication, diagnosis, or medical services claims were used as triggers to identify patients. For the calculation of VURs in patients with chronic diseases, we identified those by triggers in the given season only (Model 1) and in the given and previous seasons (Model 2). Regression analysis was used to identify factors associated with vaccination status.

Results

Data from 214,668 individual patients were analyzed. VURs over all seasons ranged from 18.4% to 19.8%. Most patients with chronic diseases were identified with the medication trigger, and we found no clinical significant differences in VURs comparing both models. Having a chronic disease, age, male gender, and regular health care provider visits were associated with increased odds of being vaccinated.

Conclusions

VURs were below the recommended thresholds, and our analysis highlighted the need for efforts to increase VURs. We assessed the identification of chronic diseases by medication claims and the calculation of VURs based on data of the given season only as an effective approach to conduct vaccination surveillance. Claims data-based surveillance may complete the national surveillance.

Falling incidence of Parkinson's disease in Germany

BACKGROUND AND PURPOSE

Idiopathic Parkinson's disease (IPD) is a progressive neurodegenerative disorder that is strongly associated with age. The aim of the present study was to describe current sex- and age-specific trends and regional differences in the incidence of IPD diagnosed in older people in Germany.

METHODS

This study was based on nationwide outpatient claims and drug prescription data from the German Statutory Health Insurance, covering approximately 87% of the general population. We conducted a cohort study in patients aged 50 years or older with observation time of at least 4 years. To assess the robustness of nationwide annual IPD incidence trends from 2013 to 2019, three case definitions with varying levels of stringency regarding coded outpatient diagnoses and drug prescriptions were applied.

RESULTS

In 2019, the population at risk comprised 30,575,726 persons. Using the primary and most specific case definition, annual age- and sex-standardized cumulative IPD incidence decreased stepwise from 137 (2013) to 106 (2019) new cases per 100,000 persons. The decline in incidence was seen in both sexes, in all age groups and in the majority of German regions. The relative decrease (2013-2019) in the annual age- and sex-standardized IPD incidence varied from 23% to 28% among case definitions.

CONCLUSION

Our findings indicate a nationwide decline in the age- and sex-standardized incidence of IPD from 2013 to 2019 in Germany. This trend was consistent using different case definitions. Further research is needed to elucidate the factors underlying this trend.

An inverse probability weighted regression method that accounts for right-censoring for causal inference with multiple treatments and a binary outcome

Comparative effectiveness research often involves evaluating the differences in the risks of an event of interest between two or more treatments using observational data. Often, the post-treatment outcome of interest is whether the event happens within a pre-specified time window, which leads to a binary outcome. One source of bias for estimating the causal treatment effect is the presence of confounders, which are usually controlled using propensity score-based methods. An additional source of bias is right-censoring, which occurs when the information on the outcome of interest is not completely available due to dropout, study termination, or treatment switch before the event of interest. We propose an inverse probability weighted regression-based estimator that can simultaneously handle both confounding and right-censoring, calling the method CIPWR, with the letter C highlighting the censoring component. CIPWR estimates the average treatment effects by averaging the predicted outcomes obtained from a logistic regression model that is fitted using a weighted score function. The CIPWR estimator has a double robustness property such that estimation consistency can be achieved when either the model for the outcome or the models for both treatment and censoring are correctly specified. We establish the asymptotic properties of the CIPWR estimator for conducting inference, and compare its finite sample performance with that of several alternatives through simulation studies. The methods under comparison are applied to a cohort of prostate cancer patients from an insurance claims database for comparing the adverse effects of four candidate drugs for advanced stage prostate cancer.

Analysis of post-extraction bleeding in patients taking antithrombotic therapy using data from the longevity improvement and fair evidence study

BACKGROUND

Several studies have investigated post-extraction bleeding in patients on antithrombotic therapy, but most included a small sample size.

OBJECTIVE

This study aimed to analyse post-extraction bleeding in patients on antithrombotic therapy using data from a large database.

MATERIALS AND METHODS

Claims data of National Health Insurance and Late-Stage Elderly Healthcare System enrollees who underwent tooth extraction between October 2014 and March 2019 (n = 107 767) in a large multiregional cohort study (Longevity Improvement and Fair Evidence study) were included. Antithrombotic therapy was determined based on the drug codes used at the time of tooth extraction (classified into six groups: no antithrombotic, single antiplatelet, dual antiplatelet, Direct Oral Anticoagulant, warfarin and combined antiplatelet and anticoagulant therapies). The outcome was defined as the presence of "post-extraction bleeding" as a receipt disease name in the same month as tooth extraction. To examine the association between antithrombotic therapy and post-extraction bleeding in detail, multiple logistic regression analysis was performed with post-extraction bleeding as the objective variable; each antithrombotic therapy as the explanatory variable; and age, sex and comorbidities as adjustment variables.

RESULTS

Antithrombotic therapy was administered in 14 343 patients (13.3%), and post-extraction bleeding was observed in 419 patients (0.4%). The rate of post-extraction bleeding was significantly lower in the no antithrombotic therapy and single antiplatelet groups than that in the other groups (odds ratio: 2.00-9.02).

CONCLUSION

The frequency of post-extraction bleeding is high in patients on anticoagulation or dual antithrombotic therapy. Therefore, careful preparation before extraction is necessary in these patients.

Healthcare Costs of Clostridioides difficile Infection in Commercially Insured Younger Adults

In a US adult population aged <65 years, attributable costs due to Clostridioides difficile infection (CDI) were highest in persons with hospital onset and lowest in those with community-associated CDI treated outside a hospital. The economic burden of CDI in younger adults underscores the need for additional CDI-preventive strategies.

Pragmatic guidance for embedding pragmatic clinical trials in health plans: Large simple trials aren't so simple

BACKGROUND

There are unique opportunities related to the design and conduct of pragmatic trials embedded in health insurance plans, which have longitudinal data on member/patient demographics, dates of coverage, and reimbursed medical care, including prescription drug dispensings, vaccine administrations, behavioral healthcare encounters, and some laboratory results. Such trials can be large and efficient, using these data to identify trial-eligible patients and to ascertain outcomes.

METHODS

We use our experience primarily with the National Institutes of Health Pragmatic Trials Collaboratory Distributed Research Network, which comprises health plans that participate in the US Food & Drug Administration's Sentinel System, to describe lessons learned from the conduct and planning of embedded pragmatic trials.

RESULTS

Information is available for research on more than 75 million people with commercial or Medicare Advantage health plans. We describe three studies that have used or plan to use the Network, as well as a single health plan study, from which we glean our lessons learned.

CONCLUSIONS

Studies that are conducted in health plans provide much-needed evidence to drive clinically meaningful changes in care. However, there are many unique aspects of these trials that must be considered in the planning, implementation, and analytic phases. The type of trial best suited for studies embedded in health plans will be those that require large sample sizes, simple interventions that could be disseminated through health plans, and where data available to the health plan can be leveraged. These trials hold potential for substantial long-term impact on our ability to generate evidence to improve care and population health.

German Claims Data for Real-World Research: Content Coverage Evaluation in OMOP CDM

Research on real-world data is becoming increasingly important. The current restriction to clinical data in Germany limits the view of the patient. To gain comprehensive insights, claims data can be added to the existing knowledge. However, standardized transfer of German claims data into OMOP CDM is currently not possible. In this paper, we conducted an evaluation regarding the coverage of source vocabularies and data elements of German claims data in OMOP CDM. We point out the need to extend vocabularies and mappings to support research on German claims data.

Sensitivity of Medicaid Claims Data for Identifying Opioid Use Disorder in Patients Admitted to 6 New York City Public Hospitals

OBJECTIVES

Behavioral health diagnoses are frequently underreported in administrative health data. For a pragmatic trial of a hospital addiction consult program, we sought to determine the sensitivity of Medicaid claims data for identifying patients with opioid use disorder (OUD).

METHODS

A structured review of electronic health record (EHR) data was conducted to identify patients with OUD in 6 New York City public hospitals. Cases selected for review were adults admitted to medical/surgical inpatient units who received methadone or sublingual buprenorphine in the hospital. For cases with OUD based on EHR review, we searched for the hospitalization in Medicaid claims data and examined International Classification of Diseases, Tenth Revision discharge diagnosis codes to identify opioid diagnoses (OUD, opioid poisoning, or opioid-related adverse events). Sensitivity of Medicaid claims data for capturing OUD hospitalizations was calculated using EHR review findings as the reference standard measure.

RESULTS

Among 552 cases with OUD based on EHR review, 465 (84.2%) were found in the Medicaid claims data, of which 418 (89.9%) had an opioid discharge diagnosis. Opioid diagnoses were the primary diagnosis in 49 cases (11.7%), whereas in the remainder, they were secondary diagnoses.

CONCLUSION

In this sample of hospitalized patients receiving OUD medications, Medicaid claims seem to have good sensitivity for capturing opioid diagnoses. Although the sensitivity of claims data may vary, it can potentially be a valuable source of information about OUD patients.

Real-world treatment patterns and healthcare resource utilization among migraine patients: A German claims database analysis

AIMS

Despite migraine being one of the most common neurological diseases, affected patients are often not effectively treated. This analysis describes the burden of migraine in Germany and assesses real-world treatment patterns and healthcare resource utilization (HCRU) of preventive-treated migraine patients from the perspective of the Statutory Health Insurance.

METHODS

A retrospective analysis was conducted using InGef Research Database claims data from 2018-2019. Migraine patients were stratified into cohorts by acute and preventive treatment status. Patients on preventive treatment were further stratified according to type of prophylaxis received. Disease burden in preventively treated migraine patients was reported via treatment patterns, pathways, and comorbidities. HCRU was assessed through outpatient provider visits, hospitalizations, and sick leave.

RESULTS

160,164 adult migraine patients were identified, of which 55,378 (34.6%) were prescribed preventive treatment with conventional (n = 25,984, 46.9%), calcitonin gene-related peptide monoclonal antibody (CGRP mAb) (n = 613, 1.1%), or off-label therapies (n = 28,781, 52.0%). 936 (1.7%) patients received Botulinum Neurotoxin Type A (BoNTA). CGRP mAb-treated patients had a high rate of triptan prescriptions (2018: 95.5%; 2019: 88.9%), migraine-related hospitalizations (2018: 33.0%; 2019: 21.0%), and sick leave (2018: 26.8%; 2019: 22.5%). A high proportion of CGRP mAb- and BoNTA-treated patients was diagnosed with abdominal and pelvic pain (34.3% and 36.2%) and low back pain (34.1% and 35.3%). These patients also showed a high prevalence of depressive episodes (49.1% and 50.1%) and chronic pain disorders (37.5% and 32.9%).

LIMITATIONS

This study focused on descriptive analyses which do not allow for assessment of causality when comparing treatment groups.

CONCLUSIONS

Disease burden was high in patients receiving CGRP mAbs suggesting that patients treated preventively with CGRP mAbs shortly after product launch in Germany were severely affected chronic migraine patients. The same may be true for patients receiving BoNTA who also showed an increased disease burden.

Prescription patterns and therapeutic effects of second-line drugs in Japanese patients with type 2 diabetes mellitus: Analysis of claims data for metformin and dipeptidyl peptidase-4 inhibitors as the first-line hypoglycemic agents

BACKGROUND

Dipeptidyl peptidase-4 inhibitors (DPP4is) and metformin are the most frequently prescribed first-line drugs for Japanese patients with type 2 diabetes (T2D). We investigated the risk of cardiovascular events by second-line treatment type in these patients.

RESEARCH DESIGN AND METHODS

Patients with T2D, prescribed either metformin or DPP4i as a first-line drug, were identified in claims data from Japanese acute care hospitals. The primary and secondary outcomes were cumulative risks of MI or stroke and of death, respectively, from second-line treatment initiation.

RESULTS

Patients prescribed first-line metformin or DPP4i was 16,736 and 74,464, respectively. In patients receiving first-line DPP4i, the death incidence was lower in those receiving second-line metformin than in those receiving second-line sulfonylurea (p < 0.001), whereas the primary outcome was not significantly different. No significant differences were observed for either outcome when DPP4is and metformin were used as first- and second-line drugs or vice versa.

CONCLUSIONS

Metformin was suggested to have larger effect to reduce death than sulfonylurea in patients receiving first-line DPP4i. The order of first- and second-line for the DPP4i and metformin combination did not affect the outcomes. Given the nature of the study design, certain limitations, including potential under-adjustment for confounders, should be considered.

Comparative evaluation of costs and healthcare resource utilization of oral selexipag versus inhaled treprostinil or oral treprostinil in patients with pulmonary arterial hypertension

BACKGROUND

Pulmonary arterial hypertension (PAH), a rare vasculopathy progressively leading to right heart failure and death, is associated with considerable economic burden. Oral prostacyclin pathway agents (PPAs) like selexipag and treprostinil address an underlying PAH pathway, yet are often under-utilized. Data on head-to-head cost comparison of various PPAs is lacking.

METHODS

In this retrospective study using a large health claims database, we compared the per-patient-per-year (PPPY) costs and healthcare resource utilization (HRU) among PAH patients taking either oral selexipag, inhaled treprostinil or oral treprostinil in the United States between July 2015 and March 2020. Patients with ≥ 1 prescription for one of the drugs of interest, ≥ 1 in-patient pulmonary hypertension (PH) diagnosis, or ≥ 2 outpatient PH diagnoses were included in this study. Baseline differences between the three groups were adjusted using an inverse probability of treatment weighting approach. 411 patients were selected for the final study cohorts.

RESULTS

All-cause hospitalization costs were highest for oral treprostinil ($39,983) compared to oral selexipag ($20,635) and inhaled treprostinil ($16,548; p = 0.037). Total PAH-related medical costs were 40% lower for patients on oral selexipag compared to patients on oral and inhaled treprostinil ($24,351 vs. $40,398 and $40,339, respectively; p = 0.006). PAH-related outpatient visits were lowest for patients on oral selexipag (14 PPPY visits) compared to oral treprostinil (16 PPPY visits) and inhaled treprostinil (22 PPPY visits; p = 0.001).

CONCLUSIONS

Compared to oral and inhaled treprostinil, oral selexipag may incur lower medical costs and reduce PAH related outpatient visits for patients with PAH.

Polyp detection rate and cumulative incidence of post-colonoscopy colorectal cancer in Germany

Studies have shown that the quality of colonoscopy influences the incidence of post-colonoscopy colorectal cancer (PCCRC). However, data from Germany on this association are lacking. We aimed to assess cumulative incidence of PCCRC in persons undergoing colonoscopy in Germany according to the physician's polyp detection rate (PDR). Using the German Pharmacoepidemiological Research Database (GePaRD) with claims data of ~20% of the German population, we included persons with a baseline colonoscopy between 2008 and 2017 and categorized them according to the procedure at baseline (snare polypectomy, forceps polypectomy, no polypectomy). In each subgroup, we distinguished between persons examined by physicians with a PDR in the lowest quartile vs higher quartiles and described cumulative CRC incidence during follow-up. Overall, 822 715 persons examined by 1752 physicians were included. One quarter of the physicians had a PDR ≤21.8% (lowest quartile). In all subgroups, the 5-year cumulative CRC incidence was statistically significantly higher in persons examined by physicians with a PDR ≤21.8% vs >21.8%: It was 69% higher in persons with snare polypectomy (0.88% vs 0.52%), 87% higher in persons with forceps polypectomy (0.58% vs 0.31%), and 48% higher in persons without polypectomy at baseline (0.31% vs 0.21%). In conclusion, we found a substantially increased PCCRC risk in persons examined by physicians with a low PDR in Germany, irrespective of the baseline findings. Our study highlights the importance of a high-quality colonoscopy to maximize the preventive effect of colonoscopy on CRC incidence.

Big Data Analytics to Reduce Preventable Hospitalizations-Using Real-World Data to Predict Ambulatory Care-Sensitive Conditions

The purpose of this study was to develop a prediction model to identify individuals and populations with a high risk of being hospitalized due to an ambulatory care-sensitive condition who might benefit from preventative actions or tailored treatment options to avoid subsequent hospital admission. A rate of 4.8% of all individuals observed had an ambulatory care-sensitive hospitalization in 2019 and 6389.3 hospital cases per 100,000 individuals could be observed. Based on real-world claims data, the predictive performance was compared between a machine learning model (Random Forest) and a statistical logistic regression model. One result was that both models achieve a generally comparable performance with c-values above 0.75, whereas the Random Forest model reached slightly higher c-values. The prediction models developed in this study reached c-values comparable to existing study results of prediction models for (avoidable) hospitalization from the literature. The prediction models were designed in such a way that they can support integrated care or public and population health interventions with little effort with an additional risk assessment tool in the case of availability of claims data. For the regions analyzed, the logistic regression revealed that switching to a higher age class or to a higher level of long-term care and unit from prior hospitalizations (all-cause and due to an ambulatory care-sensitive condition) increases the odds of having an ambulatory care-sensitive hospitalization in the upcoming year. This is also true for patients with prior diagnoses from the diagnosis groups of maternal disorders related to pregnancy, mental disorders due to alcohol/opioids, alcoholic liver disease and certain diseases of the circulatory system. Further model refinement activities and the integration of additional data, such as behavioral, social or environmental data would improve both model performance and the individual risk scores. The implementation of risk scores identifying populations potentially benefitting from public health and population health activities would be the next step to enable an evaluation of whether ambulatory care-sensitive hospitalizations can be prevented.

Trends of Antidiabetic and Cardiovascular Diseases Medication Prescriptions in Type 2 Diabetes between 2005 and 2017-A German Longitudinal Study Based on Claims Data

BACKGROUND

With an attempt to understand possible mechanisms behind the severity-dependent development of type 2 diabetes (T2D) comorbidities, this study examines the trends of antidiabetic and cardiovascular diseases (CVD) medication prescriptions in individuals with T2D.

METHODS

The study is based on claims data from a statutory health insurance provider in Lower Saxony, Germany. The period prevalence of antidiabetic and CVD medication prescriptions was examined for the periods 2005-2007, 2010-2012, and 2015-2017 in 240,241, 295,868, and 308,134 individuals with T2D, respectively. (Ordered) logistic regression analyses were applied to examine the effect of time period on the number and prevalence of prescribed medications. Analyses were stratified by gender and three age groups.

RESULTS

The number of prescribed medications per person has increased significantly for all examined subgroups. For the two younger age groups, insulin prescriptions decreased but those of non-insulin medications increased, while both increased significantly over time for the age group of 65+ years. Except for glycosides and antiarrhythmic medications, the predicted probabilities for CVD medications increased over the examined periods, with lipid-lowering agents demonstrating the highest increase.

CONCLUSIONS

Results point towards an increase in medication prescriptions in T2D, which is in line with the evidence of the increase in most comorbidities indicating morbidity expansion. The increase in CVD medication prescriptions, especially lipid-lowering agents, could explain the specific development of severe and less severe T2D comorbidities observed in this population.

Comparative performance of distinct frailty measures among patients undergoing percutaneous left atrial appendage closure

AIMS

Frailty is associated with increased morbidity and mortality in patients undergoing left atrial appendage closure (LAAC). This study aimed to compare the performance of two claims-based frailty measures in predicting adverse outcomes following LAAC.

METHODS

We identified patients 66 years and older who underwent LAAC between October 1, 2016, and December 31, 2019, in Medicare fee-for-service claims. Frailty was assessed using the previously validated Hospital Frailty Risk Score (HFRS) and Kim Claims-based Frailty Index (CFI). Patients were identified as frail based on HFRS ≥5 and CFI ≥0.25.

RESULTS

Of the 21,787 patients who underwent LAAC, frailty was identified in 45.6% by HFRS and 15.4% by CFI. There was modest agreement between the two frailty measures (kappa 0.25, Pearson's correlation 0.62). After adjusting for age, sex, and comorbidities, frailty was associated with higher risk of 30-day mortality, 1-year mortality, 30-day readmission, long hospital stay, and reduced days at home (p < .01 for all) regardless of the frailty measure used. The addition of frailty to standard comorbidities significantly improved model performance to predict 1-year mortality, long hospital stay, and reduced days at home (Delong p-value < .001).

CONCLUSION

Despite significant variation in frailty detection and modest agreement between the two frailty measures, frailty status remained highly predictive of mortality, readmissions, long hospital stay, and reduced days at home among patients undergoing LAAC. Measuring frailty in clinical practice, regardless of the method used, may provide prognostic information useful for patients being considered for LAAC, and may inform shared decision-making in this population.

Primary care referral patterns for patients with asthma: analysis of real-world data

OBJECTIVE

To identify features related to management of patients prior to referral from primary care physicians (PCPs) to pulmonologists and allergists.

METHODS

This is an analysis of patient claims data from Symphony Health (2013-2018). To characterize referrals, a longitudinal cohort included 12 months with no asthma claims prior to the index date, followed by 36 months of observation. We also assessed a cross-sectional cohort for 12 months at the end of the observational period to characterize disease control and treatment patterns. Referral was defined as the first appearance of a claim from an allergist or pulmonologist for a patient's initial visit for asthma. Descriptive statistics were used to analyze the data.

RESULTS

The majority of patients with asthma were managed by PCPs (60%), followed by pulmonologists (16%) and allergists (8%). Forty-three percent had uncontrolled asthma. Only 8% were referred to specialists within the first 24 months after initial diagnosis, of which 76% were seen by pulmonologists and 24% by allergists. Referrals resulted in treatment change in 55%-68% of the cases. Patients who received a referral were more likely to be on oral corticosteroids (OCS) and/or have more hospitalizations/ED visits.

CONCLUSIONS

About one-third of the patients managed by PCPs received intermittent and/or chronic OCS prior to referral, which may be an indication of uncontrolled disease. The referral patterns in this analysis illustrate underutilization of specialists in the consultation and management of patients with uncontrolled asthma.

Investigating drug safety in pregnancy based on the German Pharmacoepidemiological Research Database (GePaRD): A proof-of-concept analysis on the association between valproate and spina bifida

PURPOSE

Large health-care databases are increasingly used for research on drug utilization and safety in pregnancy. For the German Pharmacoepidemiological Research Database (GePaRD), covering ~20% of the German population, algorithms have been developed to identify pregnancies, to estimate their date of onset and to link mothers to their babies. Using this methodology, we aimed to conduct a proof-of-concept analysis on the known association between valproate (VPA) exposure in early pregnancy and spina bifida in the exposed child.

METHODS

We identified all pregnancies in GePaRD between 2006 and 2016 in women aged 12 to 50 years. To each VPA dispensation of these women, an exposure period was assigned, based on the dispensation date and the number of defined daily doses in the dispensed package. A pregnancy was classified as exposed to VPA in the critical time window if this exposure period overlapped with the first 55 days of pregnancy. Risk ratios were calculated for spina bifida in live births and induced abortions comparing VPA-exposed ones to all pregnancies.

RESULTS

Overall, we identified 1 271 384 pregnancies fulfilling the inclusion criteria. Of these, 668 pregnancies (0.053%) were classified as exposed to VPA in the critical time window regarding spina bifida. An induced abortion accompanied by a diagnosis of spina bifida was observed in one of the VPA-exposed pregnancies (0.15%) and in 154 of all pregnancies (0.012%), yielding a risk ratio of 12.4 (95% confidence interval [CI]: 1.7-88.2). Out of 775 875 pregnancies ending in a live birth, 366 (0.047%) were classified as VPA exposed. A diagnosis of spina bifida was coded in 3 of 366 VPA-exposed live births (0.82%) and in 260 of all live births (0.03%), yielding a relative risk of 24.5 (95% CI: 7.9-76.0).

CONCLUSIONS

Our proof-of-concept analysis based on GePaRD showed a strong association between intrauterine exposure to VPA and occurrence of spina bifida. The results are plausible and consistent with the literature, supporting the suitability of GePaRD and the developed algorithms to conduct studies on drug safety in pregnancy.

A Machine Learning Approach to Support Urgent Stroke Triage Using Administrative Data and Social Determinants of Health at Hospital Presentation: Retrospective Study

BACKGROUND

The key to effective stroke management is timely diagnosis and triage. Machine learning (ML) methods developed to assist in detecting stroke have focused on interpreting detailed clinical data such as clinical notes and diagnostic imaging results. However, such information may not be readily available when patients are initially triaged, particularly in rural and underserved communities.

OBJECTIVE

This study aimed to develop an ML stroke prediction algorithm based on data widely available at the time of patients' hospital presentations and assess the added value of social determinants of health (SDoH) in stroke prediction.

METHODS

We conducted a retrospective study of the emergency department and hospitalization records from 2012 to 2014 from all the acute care hospitals in the state of Florida, merged with the SDoH data from the American Community Survey. A case-control design was adopted to construct stroke and stroke mimic cohorts. We compared the algorithm performance and feature importance measures of the ML models (ie, gradient boosting machine and random forest) with those of the logistic regression model based on 3 sets of predictors. To provide insights into the prediction and ultimately assist care providers in decision-making, we used TreeSHAP for tree-based ML models to explain the stroke prediction.

RESULTS

Our analysis included 143,203 hospital visits of unique patients, and it was confirmed based on the principal diagnosis at discharge that 73% (n=104,662) of these patients had a stroke. The approach proposed in this study has high sensitivity and is particularly effective at reducing the misdiagnosis of dangerous stroke chameleons (false-negative rate <4%). ML classifiers consistently outperformed the benchmark logistic regression in all 3 input combinations. We found significant consistency across the models in the features that explain their performance. The most important features are age, the number of chronic conditions on admission, and primary payer (eg, Medicare or private insurance). Although both the individual- and community-level SDoH features helped improve the predictive performance of the models, the inclusion of the individual-level SDoH features led to a much larger improvement (area under the receiver operating characteristic curve increased from 0.694 to 0.823) than the inclusion of the community-level SDoH features (area under the receiver operating characteristic curve increased from 0.823 to 0.829).

CONCLUSIONS

Using data widely available at the time of patients' hospital presentations, we developed a stroke prediction model with high sensitivity and reasonable specificity. The prediction algorithm uses variables that are routinely collected by providers and payers and might be useful in underresourced hospitals with limited availability of sensitive diagnostic tools or incomplete data-gathering capabilities.

Population-Based, Spatial Analysis of Specialised Ambulatory Palliative Care in Mecklenburg-Western Pomerania, Germany, on the Basis of Reimbursement Data

In rural areas, healthcare providers, patients and relatives have to cover long distances. For specialised ambulatory palliative care (SAPV), a supply radius of max. 30 km is recommended. The aim of this study was to analyse whether there are regional disparities in the supply of SAPV and whether it is associated with the distance between the SAPV team's site and the patient's location. Therefore, anonymised data of the Association of Statutory Health Insurance Physicians of the Federal State of Mecklenburg-Western Pomerania (M-V) were retrospectively analysed for the period of 2014-2017. Identification as a palliative patient was based on palliative-specific items from the ambulatory reimbursement catalogue. In total, 6940 SAPV patients were identified; thereof, 48.9% female. The mean age was 73.3 years. For 28.3% of the identified SAPV patients (n = 1961), the SAPV teams had a travel distance of >30 km. With increasing distance, the average number of treatment days per patient increased. It was found that there are regional disparities in the provision of SAPV services in M-V and that local structures have an important impact on regional supply patterns. The distance between the SAPV team's site and the patient's location is not the only determining factor; other causes must be considered.

[Exploring for Drug Development Seeds for Children Using a Large Claims Data-Survey for Powderization or Decapsulation of Oral Tablets or Capsule Formula in Children]

Medical big data is accumulated numerous medical related data day by day. These data may have tips for new approach for drug development. Authors tried to find drug-development-needs in children using medical big data analysis with prescription survey. Medical big data were provided from JMDC (Japan Medical Data Centre) Inc. about 3 million participants between January 2005 and June 2017. In these, we identified randomly identified 22787 participants from 466701 participants who are aged ≤11 years. In these participants, 9644 were administered "capsule," "tablet," "orally distegrating tablet," "controlled release tablet/capsule" or "enteric coated tablet" formula drugs. In these, 514 were administered these as powderization or decapsulation. Sixty components administered in 145 participants (28.2%) are not marketed for pediatric formula. On the other hands, 92 components administered in 369 participants (71.8%) are decapsulation or powderization, though pediatric formulas are marketed. These 152 components may have a development seeds for children. In conclusion, prescription survey using medical big data may partially resolve the drug-development-need in pediatrics because by using medical big data will leads low biased data depending each institution.

Outcomes After Lumbar Disk Herniation Surgery in the Dutch Population

STUDY DESIGN

Retrospective cohort study.

OBJECTIVE

There is only limited data on the outcome of primary surgery of lumbar disk herniation (LDH) in Dutch patients. The objective of this study is to describe undesirable outcomes after primary LDH.

METHODS

The National Claims Database (Vektis) was searched for primary LDH operations performed from July 2015 until June 2016, for reoperations within 18 months, prescription of opioids between 6 to 12 months and nerve root block within 1 year. A combined outcome measure was also made. Group comparisons were analyzed with the Student's t-test.

RESULTS

Primary LDH surgery was performed in 6895 patients in 70 hospitals. Weighted mean of reoperations was 7.3%, nerve root block 6.7% and opioid use 15.6%. In total, 23.0% of patients had one or more undesirable outcomes after surgery. The 95% CI interval exceeded the 50% incidence line for 14 out of 26 hospitals with less than 50 surgical interventions per year. Although the data suggested a volume effect on undesired outcomes, the t-tests between hospitals with volume thresholds of 100, 150 and 200 interventions per year did not support this (P values 0.078, 0.129, 0.114).

CONCLUSION

This unique nationwide claims-based study provides insight into patient-relevant undesirable outcomes such as reoperation, nerve root block and opioid use after LDH surgery. About a quarter of the patients had a serious complication in the first follow up year that prompted further medical treatment. There is a wide variation in complication rates between hospitals with a trend that supports concentration of LDH care.

Development and validation of novel multimorbidity indices for older adults

BACKGROUND

Measuring multimorbidity in claims data is used for risk adjustment and identifying populations at high risk for adverse events. Multimorbidity indices such as Charlson and Elixhauser scores have important limitations. We sought to create a better method of measuring multimorbidity using claims data by incorporating geriatric conditions, markers of disease severity, and disease-disease interactions, and by tailoring measures to different outcomes.

METHODS

Health conditions were assessed using Medicare inpatient and outpatient claims from subjects age 67 and older in the Health and Retirement Study. Separate indices were developed for ADL decline, IADL decline, hospitalization, and death, each over 2 years of follow-up. We validated these indices using data from Medicare claims linked to the National Health and Aging Trends Study.

RESULTS

The development cohort included 5012 subjects with median age 76 years; 58% were female. Claims-based markers of disease severity and disease-disease interactions yielded minimal gains in predictive power and were not included in the final indices. In the validation cohort, after adjusting for age and sex, c-statistics for the new multimorbidity indices were 0.72 for ADL decline, 0.69 for IADL decline, 0.72 for hospitalization, and 0.77 for death. These c-statistics were 0.02-0.03 higher than c-statistics from Charlson and Elixhauser indices for predicting ADL decline, IADL decline, and hospitalization, and <0.01 higher for death (p < 0.05 for each outcome except death), and were similar to those from the CMS-HCC model. On decision curve analysis, the new indices provided minimal benefit compared with legacy approaches. C-statistics for both new and legacy indices varied substantially across derivation and validation cohorts.

CONCLUSIONS

A new series of claims-based multimorbidity measures were modestly better at predicting hospitalization and functional decline than several legacy indices, and no better at predicting death. There may be limited opportunity in claims data to measure multimorbidity better than older methods.

Comparing Lookback Periods to Ascertain Alzheimer's Disease and Related Dementias

Claims data are a valuable resource for studying Alzheimer's disease and related dementias (ADRD). Alzheimer's disease and related dementias is often identified using a list of claims codes and a fixed lookback period of 3 years of data. However, a 1-year lookback or an approach using all-available lookback data could be beneficial based on different research questions. Thus, the purpose of this study was to compare 1-year and all-available lookback approaches to ascertaining ADRD compared to the standard 3-year approach. Using a cohort of Veterans hospitalized for heart failure (N = 373, 897), our results suggested high agreement (93% or greater) between the lookback periods. The 1-year lookback period had lower sensitivity (60%) and underestimated the prevalence of ADRD. These results suggest that 1-year and all-available lookback periods are viable approaches when using claims data.