Inventory of real-world data sources in Japan: Annual survey conducted by the Japanese Society for Pharmacoepidemiology Task Force.

PURPOSE: The Database Task Force of the Japan Society for Pharmacoepidemiology began its annual surveys of databases available for clinico and pharmacoepidemiological studies in 2010. In this report, we summarize the characteristics of the databases available in Japan based on the results of our 2021 survey to illustrate the recent developments in the infrastructure for database research in Japan. METHODS: We included 20 major databases from the academia, government, or industry that were accessible to third parties. We used a web-based questionnaire to ask the database providers about their characteristics, such as their organization, data source(s), numbers of individuals enrolled, age distribution, code(s) used, and average follow-up periods. RESULTS: We received responses from all 20 databases approached: eight hospital-based databases, six insurer-based databases, four pharmacy-based databases, and two in the "other" category. Among them, 17 contained information from medical claims, pharmacy claims, and/or Diagnosis Procedure Combination data. Most insurer databases contained health check-up data that could be attached to the claims component. Some hospital-based databases had data from electronic medical records. Most insurer-based databases collected data from the insurers of working-age employees and therefore had limited coverage of older people. Most databases coded their medication data using the Japanese reimbursement codes, and many provided Anatomical Therapeutic Chemical Classification codes. CONCLUSIONS: The number of databases available for clinico and pharmacoepidemiological research and the proportion of the population they cover are increasing in Japan. The differences in their characteristics mean that the appropriate database must be selected for a particular study purpose.

Treatment patterns and characteristics of patients with migraine: results from a retrospective database study in Japan.

BACKGROUND: Clinical characteristics and treatment practice of patients with migraine in Japan in real-world setting have not been fully investigated. We conducted a retrospective cohort study using claims database to understand the clinical practice of migraine in recent years and to characterize patients potentially not managed well by current treatment options. METHODS: Our study used data from the large claims database maintained by JMDC Inc. Patients with diagnosis of headache or migraine between January 1, 2018, and July 31, 2022, were defined as the headache cohort, and those with migraine diagnosis and prescription of migraine treatments among the headache cohort were included in the migraine cohort. In the headache cohort, characteristics of medical facilities and status of imaging tests to distinguish secondary headache were examined. Treatment patterns and characteristics of patients potentially not managed well by acute/preventive treatment were described in migraine cohort. RESULTS: In the headache cohort, 989,514 patients were included with 57.0% females and mean age of 40.3 years; 77.0% patients visited clinics (with ≤ 19 bed capacities) for their primary diagnosis, and 30.3% patients underwent imaging tests (computed tomography and/or magnetic resonance imaging). In the migraine cohort, 165,339 patients were included with 65.0% females and mean age of 38.8 years. In the migraine cohort, 95.6% received acute treatment while 20.8% received preventive treatment. Acetaminophen/non-steroidal anti-inflammatory drugs were most common (54.8%) as the initial prescription for migraine treatment followed by triptan (51.4%). First treatment prescription included preventive treatment in 15.6%, while the proportion increased to 82.2% in the fourth treatment prescription. Among patients with more than 12 months of follow-up, 3.7% had prescription patterns suggestive of risk of medication-overuse headache, and these patients were characterized by a higher percentage of females and a higher prevalence of comorbidities. CONCLUSIONS: This study revealed that approximately one-fifth of the patients with migraine visiting medical facilities use preventive drugs. The presence of potential patients at risk of medication-overuse headache and the role of clinics in migraine treatment were also described.

Real-world treatment patterns of palbociclib and blood count monitoring in patients with advanced breast cancer in Japan.

Aim: To reveal the treatment patterns of palbociclib and complete blood count (CBC) monitoring in a Japanese real-world setting. Materials & methods: Deidentified data of patients with advanced breast cancer who received palbociclib from 2017 to 2020 were examined from a Japanese claims database. Results & conclusion: We identified 1074 patients. Palbociclib was commonly prescribed as second- or later-line treatment in 2017/2018; thereafter its first-line treatment increased. Regardless of treatment lines, fulvestrant was most commonly prescribed in combination with palbociclib (57-66% in the first-third-line), and this finding differed from that in the USA. Most patients initiated palbociclib at 125 mg/day; however, over a half of patients reduced doses within the first 8 weeks. Although CBC was regularly monitored, some patients did not undergo blood tests. Early dose reduction and CBC monitoring should be performed cautiously to minimize safety concern and prevent early treatment discontinuation.

Epidemiology of respiratory syncytial virus in Japan: A nationwide claims database analysis.

BACKGROUND: Respiratory syncytial virus (RSV) is a major cause of hospitalization for bronchiolitis and pneumonia in infancy. In Japan, limited data are publicly available on RSV epidemiology and clinical characteristics among infants. METHODS: This retrospective study described RSV incidence, seasonality, patient characteristics, resource use, and clinical outcomes among Japanese children <2 years from January 2017 through December 2018. The RSV cases were identified using the Japanese Medical Data Center database. RESULTS: In the database, 9,711 and 8,509 RSV patients <2 years were identified in 2017 and 2018, respectively. Of these, 25% required hospitalization. Ninety percent of hospitalized patients did not have a known RSV risk factor. Nineteen percent of hospitalized patients experienced dehydration, and 12% had acute respiratory failure. Hospitalization lasted 1 week on average and 7% required some type of mechanical ventilation. The peak of hospitalizations occurred at 2 months. The incidence of RSV hospitalization in children <2 years was 23.2 per 1,000 person-years, which increased to 35.4 per 1,000 for infants <6 months. This age group accounted for 40% of all RSV-associated hospitalizations among children <2 years. CONCLUSIONS: Roughly one-fourth of all RSV patients <2 years were hospitalized. Ninety percent of these did not have an underlying risk condition. This underscores that RSV can cause serious disease among all young children. Three to four out of every 100 Japanese children <6 months were hospitalized for RSV, and this age group accounted for ~40% of all RSV-associated hospitalizations. Novel and broad-based RSV prevention strategies, especially those targeting young infants, are needed.

Safety risk associated with use of nonsteroidal anti-inflammatory drugs in Japanese elderly compared with younger patients with osteoarthritis and/or chronic low back pain: A retrospective database study.

PURPOSE: This study aimed to assess the safety risks associated with using nonsteroidal anti-inflammatory drugs (NSAIDs) in elderly patients (≥65 years) compared with younger patients (<65 years) with osteoarthritis (OA) and/or chronic low back pain (CLBP). METHODS: A retrospective analysis was conducted on anonymized claims data of patients prescribed NSAIDs for OA and/or CLBP from 2009 to 2018 using hospital-based administrative database-Medical Data Vision (MDV). The key outcome was the incidence of developing gastrointestinal (GI), renal, and acute myocardial infarction (AMI) that are well-known events associated with NSAID use. RESULTS: Of 288,715 patients included, 23.7%, 60.5%, and 15.8% had OA, CLBP, or both, respectively. Elderly patients used non-oral NSAIDs more frequently than oral NSAIDs (57.8% and 38.7%, respectively), whereas younger patients showed comparable use (50.7% and 52.8%, respectively). The incidence of events per 10,000 person-years (95% CI) was higher in the elderly than in younger patients: GI, 29.68(27.67-31.68) vs. 16.61(14.60-18.63); renal, 124.77(120.56-128.99) vs. 39.88(36.72-43.03); and AMI, 27.41(25.48-29.35) vs. 10.90(9.27-12.53), respectively. After adjusting for covariates, the increase in risk for these events was seen in patients >70 years compared with younger patients (18-30 years) and was remarkable in patients >80 years with 2-fold, 10-fold, and 7-fold higher risk for developing GI, renal, and AMI events, respectively. CONCLUSION: Risk for developing NSAID-associated events was higher in the elderly; particularly, renal and AMI events that remarkably increased in patients >80 years. To reduce them, NSAIDs should be prescribed at the lowest effective dose for the shortest duration possible.

A Retrospective Database Study of Gastrointestinal Events and Medical Costs Associated with Nonsteroidal Anti-Inflammatory Drugs in Japanese Patients of Working Age with Osteoarthritis and Chronic Low Back Pain.

CONTEXT: The real-world burden of gastrointestinal (GI) events associated with the use of nonsteroidal anti-inflammatory drugs (NSAIDs) in Japanese patients with osteoarthritis (OA) and/or chronic low back pain (CLBP) remains unreported. OBJECTIVE: To assess the incidence and economic burden of NSAID-induced GI events by using data from large-scale real-world databases. METHODS: We used the Japanese Medical Data Center database to retrospectively evaluate anonymized claims data of medical insurance beneficiaries employed by middle- to large-size Japanese companies who were prescribed NSAIDs for OA and/or CLBP between 2009 and 2018. RESULTS: Overall, 180,371 patients were included in the analysis, of whom 32.9% had OA, 53.8% had CLBP, and 13.4% had both OA and CLBP. NSAIDs were administered as first-line analgesics to 161,152 (89.3%) of the patients in the sample, in oral form to 90.3% and as topical patches to 80.4%. A total of 65.1% used combined oral/topical patches. Of the 21.0% of patients consistently using NSAIDs (percentage of days supplied ≥70%), 54.5% received patches. A total of 51.5% patients used NSAIDs for >1 to ≤6 months. The incidence of GI events was 9.97 per 10,000 person-years (95% confidence interval: 8.92-11.03). The risk of developing GI events was high in elderly patients and patients with comorbidities and remained similar for patients receiving oral vs. topical NSAIDs. Longer treatment duration and consistent NSAID use increased the risk of GI events. The cost (median [interquartile range]) of medications (n = 327) was US$ 80.70 ($14.10, $201.40), that of hospitalization (n = 33) was US$ 2,035.50 ($1,517.80, $2,431.90), and that of endoscopic surgery (n = 52) was US$ 418.20 ($418.20, $418.20). CONCLUSION: NSAID-associated GI toxicity imposes a significant health and economic burden on patients with OA and/or CLBP, irrespective of whether oral or topical NSAIDs are used.

Efficacy of intravenous methylprednisolone pulse therapy in patients with multiple sclerosis and neuromyelitis optica.

BACKGROUND: No large-scale studies have compared the efficacy of intravenous methylprednisolone pulse therapy (IVMP) for multiple sclerosis (MS) and neuromyelitis optica (NMO). OBJECTIVE: To explain differences in treatment responses of MS and NMO patients to IVMP. METHODS: Changes in neurological symptoms/signs and Expanded Disability Status Scale (EDSS) scores before and within 1 week of IVMP completion were obtained in 2010 at 28 institutions, and retrospectively collated from 271 MS (478 courses) and 73 NMO (118 courses) cases. RESULTS: In MS patients, decreased EDSS score was significant after the first (-0.8 ± 0.9), second (-0.7 ± 0.9), and third (-0.7 ± 0.8) courses (p < 0.05), but not after the fourth (-0.3 ± 0.7) and fifth (-0.5 ± 0.6). However, decreased EDSS score was only significant after the first course (-0.5 ± 1.5, p < 0.05) in NMO patients. EDSS score was significantly decreased in MS compared with NMO patients at the first course (p < 0.05), but not thereafter. Model analysis for EDSS score improvement at the first course, adjusting for covariates, showed significantly greater decreases in MS compared with NMO patients (p < 0.05). CONCLUSION: IVMP is effective in MS from the first to third courses, and in NMO at the first course. Additionally, IVMP is more efficacious in MS than NMO patients, even at the first course.

Safety, efficacy and prognostic analyses of sunitinib in the post-marketing surveillance study of Japanese patients with gastrointestinal stromal tumor.

OBJECTIVE: This study was conducted to expand the sunitinib safety database in Japanese imatinib-resistant/-intolerant gastrointestinal stromal tumor patients. Retrospective analyses investigated common adverse events as potential prognostic markers. METHODS: Four hundred and seventy patients who received sunitinib between June 2008 and November 2009 were analyzed for safety, progression-free survival and overall survival; 386 for objective response rate; 88% received sunitinib on Schedule 4/2 starting at 50 mg/day. RESULTS: No unexpected safety issues occurred. Grade ≥ 3 adverse events occurred in 70%, most commonly thrombocytopenia (33%), neutropenia (22%) and leukopenia (15%). Objective response rate was 20% (95% confidence interval 16-24). Median progression-free survival was 22.4 weeks (95% confidence interval, 21.7-24.0). The overall survival rate at 24 weeks was 91% (95% confidence interval, 88-94). Higher relative dose intensity (≥70 vs. <70%) during the first 6 weeks and better Eastern Cooperative Oncology Group performance status (0 vs. ≥1) were associated with longer progression-free survival (24.0 vs. 20.1 weeks; P = 0.011; and 24.1 vs. 16.9 weeks; P < 0.001) and higher 24-week overall survival rate (94 vs. 83%; P < 0.001; and 96 vs. 83%; P < 0.001). Increased progression-free survival and overall survival rates were associated with specific adverse events. Cox proportional hazard modeling adjusted for relative dose intensity and performance status established hand-foot syndrome (hazard ratio = 0.636; 95% confidence interval, 0.456-0.888) and leukopenia (hazard ratio = 0.683; 95% confidence interval, 0.492-0.948) occurring within 12 weeks were significantly correlated with increased progression-free survival. CONCLUSION: Sunitinib showed good efficacy and tolerable safety. Factors associated with greater efficacy were relative dose intensity, performance status and specific early adverse events.

Cost-effectiveness analysis of 20-valent pneumococcal conjugate vaccine for routine pediatric vaccination programs in Japan.

BACKGROUND: The Japanese National Immunization Program currently includes the pediatric 13 valent pneumococcal conjugate vaccine (PCV13) to prevent pneumococcal infections. We aimed to evaluate the cost-effectiveness of 20-valent PCV (PCV20) as a pediatric vaccine versus PCV13. METHODS: A decision-analytic Markov model was used to estimate expected costs, quality-adjusted life-years (QALYs), and prevented cases and deaths caused by invasive pneumococcal disease, pneumonia, and acute otitis media over a ten-year time horizon from the societal and healthcare payer perspectives. RESULTS: PCV20 was dominant, i.e. less costly and more effective, over PCV13 (gained 294,599 QALYs and reduced Japanese yen [JPY] 352.6 billion [2.6 billion United States dollars, USD] from the societal perspective and JPY 178.9 billion [USD 1.4 billion] from the payer perspective). Sensitivity and scenario analyses validated the robustness of the base scenario results. When comparing PCV20 with PCV13, the threshold analysis revealed an incremental cost-effectiveness ratio that was within the threshold value (JPY 5 million/QALY) at a maximum acquisition cost of JPY 74,033 [USD 563] (societal perspective) and JPY 67,758 [USD 515] (payer perspective). CONCLUSIONS: As a pediatric vaccine, PCV20 was dominant over PCV13 regardless of the study perspective.

Cost-Effectiveness Analysis of Maternal Respiratory Syncytial Virus Vaccine in Protecting Infants from RSV Infection in Japan.

INTRODUCTION: Respiratory syncytial virus (RSV) is one of the major causes of respiratory tract infections among children. Until recently, the monoclonal antibody palivizumab was the only RSV prophylaxis available in Japan. In 2024, the bivalent RSV prefusion F protein-based (RSVpreF) vaccine was approved for the prevention of RSV infection in infants by active immunization of pregnant women. In this study, we assessed the cost-effectiveness of a combined strategy of RSVpreF vaccine and palivizumab in Japanese setting. METHODS: Using a Markov model, we evaluated prevented cases and deaths of medically attended RSV infections from birth to age 11 months for each of the three healthcare settings: inpatient (hospitalization), emergency department visits, and outpatient visits. Incremental cost-effectiveness ratios (ICERs) were calculated from economic outcomes (intervention costs, medication costs, and productivity losses) and quality-adjusted life years (QALYs). Further, we calculated the maximum price of RSVpreF vaccine within which the program would be cost-effective. RESULTS: In comparison with the current prophylaxis (palivizumab alone), a combined prophylaxis of year-round RSVpreF vaccination of pregnant women and palivizumab prescription for premature infants born in < 32 weeks gestational age (wGA) and all infants with high risk prevented 14,382 medically attended cases of RSV (hospitalization, 7490 cases; emergency department, 2239 cases; outpatient, 4653 cases) and 7 deaths, respectively. From a healthcare payer perspective, when the price of RSVpreF vaccine was equal to or less than ¥23,948 (US $182), a combination prophylaxis was cost-effective under the ICER threshold of ¥5 million per QALY. The other combination prophylaxis of year-round RSVpreF vaccination and palivizumab prescription of premature born in < 32 wGA regardless of risk in infants was a dominant strategy (more effective and less costly). CONCLUSION: A combined prophylaxis of year-round RSVpreF vaccine and palivizumab could be a cost-effective strategy to protect neonates throughout the infant stage (< 1 years old) in Japan.

Safety and pharmacokinetics of PF-04360365 following a single-dose intravenous infusion in Japanese subjects with mild-to-moderate Alzheimer's disease: a multicenter, randomized, double-blind, placebo-controlled, dose-escalation study.

OBJECTIVE: PF-04360365 is a humanized IgG(2)Δa anti-amyloid ß (Aß) antibody designed to improve outcome in Alzheimer's disease (AD). Single doses of 0.1 - 10 mg/kg were safe and well tolerated in Western (mostly Caucasian) subjects with mild-to-moderate AD. This Phase 1, multicenter, randomized, double-blind, dose-escalation study was the first to evaluate the safety, pharmacokinetics, pharmacodynamics, and immunogenicity of PF-04360365 in Japanese subjects. MATERIALS AND METHODS: 30 subjects with mild-to-moderate AD were enrolled. In each cohort, 3 subjects received PF-04360365 (0.1, 0.5, 1, 5, or 10 mg/kg) and 1 subject received placebo as a single 2-hour intravenous infusion. Subjects were monitored as inpatients for 24 hours and then as outpatients for 1 year. RESULTS: All subjects completed the study. There were no serious or National Cancer Institute Common Terminology Criteria for Adverse Events grade ≥ 3 adverse events, hypersensitivity reactions, or antidrug antibodies. No clinical or MRI evidence of brain microhemorrhage, cerebral edema, or encephalitis was observed. PF-04360365 plasma concentrations increased with dose, and pharmacokinetics were consistent with a small steady-state volume of distribution, slow clearance, and long elimination half-life. Cerebrospinal fluid (CSF):plasma ratios were < 0.5%. Plasma Aß species showed dose-dependent increases in C(max) and AUC(∞), but CSF biomarkers did not differ clearly between treatment arms. CONCLUSIONS: PF-04360365 was safe and well tolerated in Japanese subjects. Pharmacokinetics and plasma pharmacodynamic responses in Japanese subjects were comparable to those in Western subjects. *No longer affiliated with Pfizer.

Optimal timing for interim analyses in clinical trials.

In clinical trials, interim analyses are often performed before the completion of the trial. The intention is to possibly terminate the trial early or adjust the sample size. The time of conducting an interim analysis affects the probability of the early termination and the number of subjects enrolled until the interim analysis. This influences the expected total number of subjects. In this study, we examine the optimal time for conducting interim analyses with a view to minimizing the expected total sample size. It is found that regardless of the effect size, the optimal time of one interim analysis for the early termination is approximately two-thirds of the planned observations for the O'Brien-Fleming type of spending function and approximately half of the planned observations for the Pocock type when the subject enrollment is halted for the interim analysis. When the subject enrollment is continuous throughout the trial, the optimal time for the interim analysis varies according to the follow-up duration. We also consider the time for one interim analysis including the sample size adjustment in terms of minimizing the expected total sample size.

Real-world treatment patterns of subsequent therapy after palbociclib in patients with advanced breast cancer in Japan.

PURPOSE: The optimal treatment following endocrine therapy (ET) plus a cyclin-dependent kinase 4/6 inhibitor (CDK4/6i) has not been established. We aimed to investigate treatment patterns and time to treatment failure (TTF) of subsequent therapy after palbociclib in a Japanese real-world setting. METHDS: This retrospective observational study used de-identified data of patients with advanced breast cancer treated with palbociclib, using a nationwide claims database (April 2008 to June 2021). Measures included the type of subsequent therapies after palbociclib (endocrine-based therapy: ET alone, ET + CDK4/6i, and ET + mammalian target of rapamycin inhibitor [mTORi]; chemotherapy; chemotherapy + ET; and others) and their TTFs. The median TTF and 95% confidence interval (CI) were estimated using the Kaplan-Meier method. RESULTS: Of 1170 patients treated with palbociclib, 224 and 235 received subsequent therapies after first- and second-line palbociclib treatment, respectively. Among them, 60.7% and 52.8% were treated with endocrine-based therapies as first subsequent therapy, including ET + CDK4/6i (31.2% and 29.8%, respectively). The median TTF (95% CI) of ET alone, ET + CDK4/6i, and ET + mTORi as first subsequent therapy after first-line palbociclib were 4.4 (2.8-13.7), 10.9 (6.5-15.6), and 6.1 (5.1-7.2) months, respectively. No apparent relationship between the treatment duration of prior ET + palbociclib and subsequent abemaciclib was observed. CONCLUSION: This real-world study revealed that one-third of the patients received sequential CDK4/6i after ET + palbociclib, and treatment duration of ET + CDK4/6i following ET + palbociclib was the longest among the treatment options. Further data are awaited to determine whether ET + targeted therapy with CDK4/6i and mTORi provides acceptable treatment options following ET + palbociclib.

Incidence rates for hospitalized infections, herpes zoster, and malignancies in patients with ulcerative colitis in Japan: an administrative health claims database analysis.

BACKGROUND/AIMS: Patients with ulcerative colitis (UC) are at an increased risk of certain infections and malignancies compared with the general population. Incidence rates (IRs) of hospitalized infections, herpes zoster (HZ), and malignancies in patients with UC, stratified by treatment, in Japan were estimated. METHODS: This retrospective study identified patients with UC treated with corticosteroids, immunosuppressants, or tumor necrosis factor inhibitors (TNFi) from 2 administrative databases (Japan Medical Data Center [JMDC] and Medical Data Vision [MDV]). IRs (unique patients with events per 100 patient-years) were estimated for hospitalized infections, HZ, and malignancies, between June 2010 and May 2018. RESULTS: Among 6,033 MDV patients with UC receiving corticosteroids, immunosuppressants, or TNFi, IRs (95% confidence intervals) were: hospitalized infections, 1.73 (1.52-1.93); HZ, 1.00 (0.85-1.16), and malignancies, 1.48 (1.29-1.66). Among 958 JMDC patients with UC receiving corticosteroids, immunosuppressants, or TNFi, IRs (95% confidence intervals) were: HZ, 1.82 (1.27-2.37) and malignancies, 1.35 (0.87-1.82). In both cohorts, IRs of malignancies were generally similar among patients receiving immunosuppressants, TNFi, or combination therapy (immunosuppressants and TNFi); this was also true for IRs of hospitalized infections and HZ in the MDV cohort. IRs of hospitalized infections, HZ, and malignancies were higher in patients receiving calcineurin inhibitors compared with immunosuppressants or TNFi, in both cohorts. CONCLUSIONS: IRs of hospitalized infections, HZ, and malignancies among patients with UC were generally similar regardless of UC treatment, except for calcineurin inhibitors.

Developing Artificial Intelligence Models for Extracting Oncologic Outcomes from Japanese Electronic Health Records.

INTRODUCTION: A framework that extracts oncological outcomes from large-scale databases using artificial intelligence (AI) is not well established. Thus, we aimed to develop AI models to extract outcomes in patients with lung cancer using unstructured text data from electronic health records of multiple hospitals. METHODS: We constructed AI models (Bidirectional Encoder Representations from Transformers [BERT], Naïve Bayes, and Longformer) for tumor evaluation using the University of Miyazaki Hospital (UMH) database. This data included both structured and unstructured data from progress notes, radiology reports, and discharge summaries. The BERT model was applied to the Life Data Initiative (LDI) data set of six hospitals. Study outcomes included the performance of AI models and time to progression of disease (TTP) for each line of treatment based on the treatment response extracted by AI models. RESULTS: For the UMH data set, the BERT model exhibited higher precision accuracy compared to the Naïve Bayes or the Longformer models, respectively (precision [0.42 vs. 0.47 or 0.22], recall [0.63 vs. 0.46 or 0.33] and F1 scores [0.50 vs. 0.46 or 0.27]). When this BERT model was applied to LDI data, prediction accuracy remained quite similar. The Kaplan-Meier plots of TTP (months) showed similar trends for the first (median 14.9 [95% confidence interval 11.5, 21.1] and 16.8 [12.6, 21.8]), the second (7.8 [6.7, 10.7] and 7.8 [6.7, 10.7]), and the later lines of treatment for the predicted data by the BERT model and the manually curated data. CONCLUSION: We developed AI models to extract treatment responses in patients with lung cancer using a large EHR database; however, the model requires further improvement.

The use of artificial intelligence (AI) to derive health outcomes from large electronic health records is not well established. Thus, we built three different AI models: Bidirectional Encoder Representations from Transformers (BERT), Naïve Bayes, and Longformer to serve this purpose. Initially, we developed these models based on data from the University of Miyazaki Hospital (UMH) and later improved them using the Life Data Initiative (LDI) data set of six hospitals. The performance of the BERT model was better than the other two, and it showed similar results when it was applied to the LDI data set. The Kaplan­Meier plots of time to progression of disease for the predicted data by the BERT model showed similar trends to those for the manually curated data. In summary, we developed an AI model to extract health outcomes using a large electronic health database in this study; however, the performance of the AI model could be improved using more training data.

Real-World Amount of Clotting Factor Concentrates Dispensed and Annual Medical Expenditures for Japanese Patients with Hemophilia B.

Introduction: Until extended half-life (EHL) factor IX (FIX) concentrates became available in Japan in 2010, patients with hemophilia B received intravenous FIX replacement therapy with standard half-life (SHL) FIX concentrates. Purpose: To investigate the amount of factor dispensed and the associated medical expenditures for the treatment of hemophilia B in the real-world clinical setting in Japan. Methods: This retrospective study comprised patients with hemophilia B (N=197) who had filled prescriptions for FIX concentrates reported in Japan's Medical Data Vision database from 2015 to 2019. Patients were included if they had 2 or more prescriptions for the same FIX concentrates within the first 6 months of the study period and the interval between prescriptions was at least 2 weeks. Results: Since 2015, there was a decrease in the proportion of patients using SHL FIX concentrates and a corresponding increase in international units of dispensed EHL FIX concentrates. Median annualized dispensed dosages (IU/kg body weight) of EHL FIX concentrates were lower than for SHL concentrates for outpatient use only. Annual total health care expenditures per patient and annual expenditures for prescribed FIX concentrates increased each year during the study period. Following a switch from an SHL to an EHL concentrate, the median amount of prescribed FIX concentrate decreased slightly, although median total health care expenditures and FIX concentrate expenditures increased. Conclusion: In the real-world setting in Japan, medical expenditures and the proportion of patients prescribed EHL FIX concentrates for the treatment of hemophilia B have increased.

Real world data and data science in medical research: present and future.

Real world data (RWD) are generating greater interest in recent times despite being not new. There are various purposes of the RWD analytics in medical research as follows: effectiveness and safety of medical treatment, epidemiology such as incidence and prevalence of disease, burden of disease, quality of life and activity of daily living, medical costs, etc. The RWD research in medicine is a mixture of digital transformation, statistics or data science, public health, and regulatory science. Most of the articles describing the RWD or real-world evidence (RWE) in medical research cover only a portion of these specializations, which might lead to an incomplete understanding of the RWD. This article summarizes the overview and challenges of the RWD analysis in medical fields from methodological perspectives. As the first step for the RWD analysis, data source of the RWD should be comprehended. The progress of the RWD is closely related to the digitization, especially of medical administrative data and medical records. Second, the selection of appropriate statistical and epidemiological methods is highly critical for an RWD analysis than those for randomized clinical trials. This is because it contains greater varieties of bias, which should be controlled by balancing the underlying risk between treatment groups. Last, the future of the RWD is discussed in terms of overcoming limited data by proxy confounders, using unstructured text data, linking of multiple databases, using the RWD or RWE for a regulatory purpose, and evaluating values and new aspects in medical research brought by the RWD.

Sample size re-estimation for survival data in clinical trials with an adaptive design.

In clinical trials with survival data, investigators may wish to re-estimate the sample size based on the observed effect size while the trial is ongoing. Besides the inflation of the type-I error rate due to sample size re-estimation, the method for calculating the sample size in an interim analysis should be carefully considered because the data in each stage are mutually dependent in trials with survival data. Although the interim hazard estimate is commonly used to re-estimate the sample size, the estimate can sometimes be considerably higher or lower than the hypothesized hazard by chance. We propose an interim hazard ratio estimate that can be used to re-estimate the sample size under those circumstances. The proposed method was demonstrated through a simulation study and an actual clinical trial as an example. The effect of the shape parameter for the Weibull survival distribution on the sample size re-estimation is presented.

Burden of Renal Events Associated with Nonsteroidal Anti-inflammatory Drugs in Patients with Osteoarthritis and Chronic Low Back Pain: A Retrospective Database Study.

INTRODUCTION: Nonsteroidal anti-inflammatory drugs (NSAIDs) have long-term benefits but are limited by side effects. We assessed the health and economic burden of renal events associated with NSAID use in patients with osteoarthritis (OA) and/or chronic low back pain (CLBP). METHODS: This retrospective, large-scale, medical claims database study of Japanese patients receiving NSAIDs for OA and/or CLBP between 2009 and 2018 assessed the incidence of renal events and effect of treatment duration, mode of administration, and usage consistency of NSAIDs. RESULTS: Of 180,371 patients, NSAIDs were prescribed as first-line analgesics in 89.3%. Incidence per 10,000 person-years (95% confidence interval [CI]) for renal events was 23.46 (21.84-25.08) and for progression of chronic kidney disease (CKD) was 267.12 (189.93-344.32). Longer treatment duration (> 1 to ≤ 3 years, risk ratio [RR] 1.32, 95% CI 1.12-1.54; P = 0.0007; > 3 to ≤ 5 years, RR: 1.38, 95% CI 1.04-1.84; P = 0.0254 vs. < 1 year) and consistent use (RR: 1.24, 95% CI 0.99-1.55; P = 0.0595) increased the risk of renal events but the latter did not reach statistical significance. The risk was similar in patients using patch/oral NSAIDs and high in elderly patients and in those with diabetes, hypertension, and other cardiovascular disease. Following a renal event, median 1-year cost of drug treatment was $27.90; hospitalization, $1779.40; and dialysis, $33,018.40. CONCLUSIONS: Risk of renal events significantly increased with prolonged and consistent NSAID use (irrespective of mode of administration), with age, and in patients with certain comorbidities. Careful NSAID use is recommended in patients with CKD and those at high risk for CKD.

Nonsteroidal anti-inflammatory drugs (NSAIDs) are commonly used in patients with osteoarthritis (OA) and/or chronic low back pain (CLBP) for pain relief but their use is limited by side effects. These side effects may include abdominal, heart, and kidney problems. This article presents the results from a large claims database study in Japan that assessed the incidence of renal events and the associated healthcare cost. Impact of NSAIDs treatment duration, mode of administration, and usage consistency on the risk of developing renal events was evaluated. Results showed high incidence of renal events and progression of chronic kidney disease. Longer treatment duration and consistent use increased the risk of developing renal events. The risk was similar in patients using patch/oral NSAIDs and high in elderly patients and those with diabetes, hypertension, and other heart diseases. The estimated cost of drug treatment, hospitalization, and dialysis was also high. The author of the study would recommend NSAIDs to be used carefully in patients at risk for (or with) chronic kidney disease.

Database Analysis on the Relationships Between Nonsteroidal Anti-inflammatory Drug Treatment Variables and Incidence of Acute Myocardial Infarction in Japanese Patients with Osteoarthritis and Chronic Low Back Pain.

INTRODUCTION: We aimed to analyze the relationships between nonsteroidal anti-inflammatory drug (NSAID) treatment variables and the incidence of acute myocardial infarction (AMI) in Japanese patients with osteoarthritis (OA) and chronic low back pain (CLBP) using the data from a large-scale, real-world database. METHODS: We retrospectively analyzed anonymized claims data from the Japanese Medical Data Center of medical insurance beneficiaries who were prescribed NSAIDs for OA and/or CLBP from 2009 to 2018. RESULTS: Of 180,371 patients, 89.3% received NSAIDs as first-line analgesics (oral, 90.3%; patch, 80.4%; other transdermal drugs, 24.0%). Incidence of AMI was 10.27 per 10,000 person-years (95% confidence interval 9.20-11.34) in the entire study population. There was a trend towards increased risk in patients using NSAIDs for more than 5 years (P = 0.0784) than in those using NSAIDs for less than 1 year. Risk of AMI significantly increased with age and comorbidities of diabetes and cardiovascular disease (CVD). The risk for AMI was similar for patients who consistently used NSAIDs compared to those using them intermittently and patients who used patch compared to oral NSAIDs. Elderly patients used NSAIDs more consistently and used NSAID patches more frequently. CONCLUSION: In Japanese patients with OA and CLBP, we saw a trend of increased risk for AMI in patients using NSAIDs for more than 5 years. Elderly patients had a higher prevalence of diabetes, hypertension, and other CVD which increased the risk of AMI. Although NSAID patches were preferred to oral NSAIDs in elderly patients, risk for AMI was similar between the two modalities. Therefore, we suggest using NSAIDs carefully, especially in elderly patients and those at risk of developing CVD.