ClinicoEconomics and Outcomes Research最新文献

Introduction: Clinical trials are critical for drug development and patient care; however, they often need more efficient trial design and patient enrolment processes. This research explores integrating machine learning (ML) techniques to address these challenges. Specifically, the study investigates ML models for two critical aspects: (1) streamlining clinical trial design parameters (like the site of drug action, type of Interventional/Observational model, etc) and (2) optimizing patient/volunteer enrolment for trials through efficient classification techniques.

Methods: The study utilized two datasets: the first, with 55,000 samples (from ClinicalTrials.gov), was divided into five subsets (10,000-15,000 rows each) for model evaluation, focusing on trial parameter optimization. The second dataset targeted patient eligibility classification (from the UCI ML Repository). Five ML models-XGBoost, Random Forest, Support Vector Classifier (SVC), Logistic Regression, and Decision Tree-were applied to both datasets, alongside Artificial Neural Networks (ANN) for the second dataset. Model performance was evaluated using precision, recall, balanced accuracy, ROC-AUC, and weighted F1-score, with results averaged across k-fold cross-validation.

Results: In the first phase, XGBoost and Random Forest emerged as the best-performing models across all five subsets, achieving an average balanced accuracy of 0.71 and an average ROC-AUC of 0.7. The second dataset analysis revealed that while SVC and ANN performed well, ANN was preferred for its scalability to larger datasets. ANN achieved a test accuracy of 0.73714, demonstrating its potential for real-world implementation in patient streamlining.

Discussion: The study highlights the effectiveness of ML models in improving clinical trial workflows. XGBoost and Random Forest demonstrated robust performance for large clinical datasets in optimizing trial parameters, while ANN proved advantageous for patient eligibility classification due to its scalability. These findings underscore the potential of ML to enhance decision-making, reduce delays, and improve the accuracy of clinical trial outcomes. As ML technology continues to evolve, its integration into clinical research could drive innovation and improve patient care.

Purpose: This study evaluates the Social Return on Investment (SROI) of implementing measures to prevent fragility fractures in postmenopausal women with osteoporosis (OP) in Spain.

Methods: A group of 13 stakeholders identified necessary actions for improving refracture prevention and assessed the investment required from the Spanish National Health System (SNHS), considering direct, indirect, and intangible costs over a one-year period. Unitary costs were sourced from scientific literature and official data, and intangible costs were estimated through surveys on women's willingness to pay for better health-related quality of life. The SROI ratio was calculated from both a social perspective (including all returns) and the SNHS perspective (including only direct and intangible costs). A sensitivity analysis evaluated the returns in worst- and best-case scenarios over three years.

Results: Stakeholders agreed on four main actions: 1) establishing fracture liaison services; 2) harmonizing clinical practice guidelines and provide training for healthcare professionals (HCPs); 3) promoting HCPs' adherence to fracture registries and 4) raising awareness of OP and fragility fractures. From the social perspective, implementing these actions would cost the SNHS €4,375,663 but yield a social return of €96,939,931 in the first year, resulting in a SROI ratio of €22.15 per euro invested (€28.69, 23.14, 24.29, and 10.70 for the four actions, respectively). From the SNHS perspective, the return would be €36,453,509 (€21,523,444 tangible), with a SROI of €8.33 (€4.92 tangible) and for the four actions: €9.99, 9.39, 8.45, and 3.79, respectively (€5.89, 5.54, 4.96 and 2.27 tangible). The investment would be lower than the return for all actions (3.49%, 4.32%, 4.12% and 9.34% of social perspective return, respectively) and scenarios.

Conclusion: According to our SROI method, implementing different actions to improve secondary fracture prevention would achieve a considerable social benefit, which, in terms of direct, indirect, and intangible costs, would far outweigh the investment.

Purpose: In cardiovascular surgeries, iron deficiency anemia and transfusion of blood products are associated with mortality and morbidity, prolonged hospital stay and poor patient outcomes. Patient blood management (PBM) is a patient-centered approach based on a 'three pillar' model that promotes optimum use of blood and blood products to improve outcomes. This study assessed the potential budget impact of implementing PBM in patients undergoing elective cardiovascular surgery in a private hospital in Turkey.

Methods: Two models were developed to estimate the hospital budget impact of PBM. The first model encompassed implementation of the first pillar of PBM, which proposes treatment of iron deficiency anemia before a surgical procedure. The second covered implementation of all three pillars of PBM. Budget impact was estimated from the number of avoided complications after treating iron deficiency anemia and reducing blood transfusions. Rates of complication (sepsis, myocardial infarction, renal failure and stroke) with and without PBM were taken from published meta-analyses. Data on 882 cardiovascular operations performed during 2020-2022 were taken from the Florence Nightingale Istanbul Hospital. The costs of treating complications were estimated by applying Turkish Social Security Institution prices to a healthcare resource utilization tool for each complication completed by experts.

Results: Results from the budget impact analysis showed that, by implementing the first pillar of PBM, the department could have avoided 30 complications and saved 4,189,802 TRY. For the second model based on implementing all three pillars of PBM, 29 complications could have been avoided by reducing the number of transfusions, with budget savings of 6,174,434 TRY. Reducing the length of hospital stay could have enabled 137 additional operations in the given period.

Conclusion: Implementation of PBM in patients undergoing elective cardiovascular surgery in private hospitals could be a budget-saving strategy in Turkey and may provide an opportunity to increase revenue.

Objective: Prostate Cancer can be treated with various formulations of Gonadotropin-Releasing Hormone Agonists (GnRHa), but cost analyses of these treatments in China are lacking. This study aims to evaluate the differences in cost and resource utilization between various formulations of GnRHa for Prostate Cancer by conducting a resource utilization assessment and cost minimization analysis.

Methods: From the perspective of society and medical healthcare, this study used the cost minimization model to generate cost and resource estimates for GnRHa drug acquisition and administration for "Current practice" and for a "Base case" scenario. In the "Base case" scenario, all of the patients who were receiving 1-monthly or 3-monthly GnRHa therapy in "Current practice" switched to a 6-monthly formulation triptorelin. Cost/Resource estimates were calculated per patient per administration and scaled to annualized population levels. Deterministic sensitivity analysis was conducted to explore the uncertainty of the model variables and applied assumptions.

Results: From a societal perspective, if all 1-monthly and 3-monthly formulations of GnRHa were switched to a 6-monthly formulation triptorelin, it is conservatively estimated that the annual societal cost could be reduced by ¥13,382,951.13, with an average annual cost savings of ¥46.53 per patient. Additionally, the 6-monthly formulation could save 3,608,973.91 hours annually, translating to an average time savings of 12.55 hours per patient, reducing treatment time by 78%. From a healthcare system perspective, if the introduction of the 6-monthly formulation of GnRHa is delayed, it would lead to an annual increase of ¥94 million in medical costs, and require an additional 64,445.96 working days for doctors and nurses. Deterministic sensitivity analysis demonstrated the model's robustness, showing the 6-monthly GnRHa remains cost-effective across various parameter changes, with drug price being the most influential factor.

Conclusion: Compared to current 1-monthly and 3-monthly formulations, the 6-monthly GnRHa can reduce the total burden associated with prostate cancer treatment.

Objective: To evaluate chronic conditions as leading predictors of economic burden over time among older adults with incident primary Merkel Cell Carcinoma (MCC) using machine learning methods.

Methods: We used a retrospective cohort of older adults (age ≥ 67 years) diagnosed with MCC between 2009 and 2019. For these elderly MCC patients, we derived three phases (pre-diagnosis, during-treatment, and post-treatment) anchored around cancer diagnosis date. All three phases had 12 months baseline and 12-months follow-up periods. Chronic conditions were identified in baseline and follow-up periods, whereas annual total and out-of-pocket (OOP) healthcare expenditures were measured during the 12-month follow-up. XGBoost regression models and SHapley Additive exPlanations (SHAP) methods were used to identify leading predictors and their associations with economic burden.

Results: Congestive heart failure (CHF), chronic kidney disease (CKD) and depression had the highest average incremental total expenditures during pre-diagnosis, treatment, and post-treatment phases, respectively ($25,004, $24,221, and $16,277 (CHF); $22,524, $19,350, $20,556 (CKD); and $21,645, $22,055, $18,350 (depression)), whereas the average incremental OOP expenditures during the same periods were $3703, $3,013, $2,442 (CHF); $2,457, $2,518, $2,914 (CKD); and $3,278, $2,322, $2,783 (depression). Except for hypertension and HIV, all chronic conditions had higher expenditures compared to those without the chronic conditions. Predictive models across each of phases of care indicated that CHF, CKD, and heart diseases were among the top 10 leading predictors; however, their feature importance ranking declined over time. Although depression was one of the leading drivers of expenditures in unadjusted descriptive models, it was not among the top 10 predictors.

Conclusion: Among older adults with MCC, cardiac and renal conditions were the leading drivers of total expenditures and OOP expenditures. Our findings suggest that managing cardiac and renal conditions may be important for cost containment efforts.

Introduction: Identification and reporting of severe adverse events (SAEs) during anesthesia care remains critical in identifying areas of improvement in perioperative patient care. Although many healthcare organizations rely on the self-reporting of SAEs, under-reporting may limit the identification of the true incidence of these events. To circumvent these barriers, many healthcare systems leverage the Electronic Medical Record (EMR) by incorporating an Anesthesia Information Management System (AIMS).

Methods: We followed the Institute for Healthcare Improvement's Model of Improvement and implemented behavioral economic-based interventions to our perioperative practice including adding a deliberation-promoting "hard stop" that required the anesthesiologists to report the occurrence or absence of a "notable event" prior to closing a patient's encounter in the EMR system.

Results: At baseline, 53% of SAEs were self-reported. The interventions resulted in a baseline shift to more than 75% self-reporting, a relative increase of 42%.

Conclusion: An increase in reporting of SAEs was achieved with simple interventions including modifications of the EMR which were done with limited financial impact or interruption in the work flow.

Background and objectives: Limited information is available regarding the distribution of increasing pharmaceutical expenditures within large representative samples of national populations globally. The aim was to investigate the distribution of pharmaceutical costs in outpatient treatment and analyze the primary characteristics of users of expensive drugs within the healthcare system of Kazakhstan.

Methods: This study utilized data from the Information System for Outpatient Drug Supply, which includes nationally representative data from all regions of Kazakhstan, covering both rural and urban populations. The key explanatory variables in this study included age, gender, number of prescribed medications, disease categories based on ICD-10 codes, and insurance coverage status. These variables were selected to capture demographic, clinical, and healthcare access factors influencing prescription drug costs. In total, 2.2 million people, who were prescribed outpatient medications were included. High-cost users (HCUs) were characterized as individuals whose prescription drug expenses ranked within the highest 5%.

Results: The distribution of pharmaceutical costs exhibits significant discrepancy, with 5% of the population receiving prescription drugs covered by the state budget and social medical insurance fund contributing to nearly three-quarters of all costs. Notably, these HCUs tended to be younger than low-cost drug users. HCUs, on average, consumed a greater quantity of medications compared to non-HCUs. Among children, the top diseases contributing to high costs were rare hereditary diseases and malignancies, while in adults, cancer and diabetes were the primary cost drivers.

Conclusion: There is a concentration of public drug program spending within a small percentage of beneficiaries with high drug costs in Kazakhstan. This discovery offers valuable insights for shaping policies tailored to this specific population, aiming to mitigate escalating costs and enhance the optimal use of medications.

Purpose: To analyze the costs of high thromboembolic risk patients who require low molecular weight heparins (LMWHs) as a thromboprophylaxis strategy.

Methods: Cost analysis was conducted to assess LMWHs (enoxaparin versus comparators: nadroparin and dalteparin) as thromboprophylaxis for hospitalized patients with high thromboembolic risk in Oncology, General or Orthopedic Surgery, and Internal Medicine services from the healthcare provider's perspective in Colombia. A decision tree was developed, and the health outcomes considered in the analysis were deep vein thrombosis, major bleeding, pulmonary thromboembolism, and chronic pulmonary hypertension. Clinical inputs were obtained from a systematic review of the literature and the economic parameters from micro-costing. Inputs were validated by three clinical experts. Costs were expressed in 2020 US dollars (USD).

Results: In a hypothetical cohort of 10,000 patients with a thromboprophylaxis use rate of 40%, the use of enoxaparin was less costly than that of dalteparin in Oncology (difference of USD 624,669), Orthopedic Surgery (difference of USD 275,829), and Internal Medicine (difference of USD 109,119) patients. For these services, using enoxaparin was more efficient than using nadroparin (cost differences of USD 654,069, USD 416,927, and USD 92,070, respectively). Sensitivity analysis showed an important influence of the number of patients undergoing thromboprophylaxis, as well as the unit cost, and the risk of events (DVT, PTE, and CTEPH).

Conclusion: Enoxaparin is the least expensive health technology for thromboprophylaxis in most of the medical contexts analyzed in Colombia due to its efficacy and the lower risk of complications than dalteparin and nadroparin.

Background: Icodec, once-weekly basal insulin, aims to simplify therapy management by reducing injection frequency for diabetic patients. The efficacy and safety of icodec were evaluated in the ONWARDS clinical development program. This study evaluates icodec economic and quality of life impact from the Italian National Healthcare System (NHS) perspective.

Materials and methods: A pharmacoeconomic study was developed to assess the once-weekly insulin icodec value, highlighting its potential to decrease needle use while improving adherence and quality of life. In the base case, a differential cost and cost-utility analysis over one year compared to once-daily insulin degludec were developed. Based on the comparison with degludec, a scenario analysis was planned between icodec and the mix of basal insulins available on the market. Economic evaluations included drug and administration costs, needles, and impact on adherence. The cost-utility analysis measured the utility associated with the weekly injection compared to the daily ones, resulting in an incremental cost-effectiveness ratio (ICER), measured as Δ€/ΔQALY (Quality Adjusted Life Years). To assess the robustness of the results, a deterministic one-way sensitivity analysis and a probabilistic sensitivity analysis were carried out.

Results: At an annual cost 25% higher than degludec, considering the economic benefits generated by the needle use reduction (-€51.10) and adherence improvement (-€54.85), once-weekly icodec grants no incremental cost and even potential savings per patient. Furthermore, icodec reported a utility advantage (0.023). It achieved a dominant incremental cost-effectiveness ratio (ICER) compared to degludec. The comparison with the mix of basal insulins also reported a cost-effectiveness profile. Sensitivity tests conducted confirmed the robustness of the findings, highlighting the key drivers of the analysis.

Conclusion: Icodec represents a new therapeutic option to simplify basal insulin treatment. It also improves the patient's management and his quality of life, without increasing the economic burden for the Italian NHS, while guaranteeing an excellent cost-effectiveness profile.