Pragmatic and Observational Research最新文献

{"title":"Shifting from Traditional to Pragmatic Randomized Controlled Trials: Insights and Lessons Learned from the Toddler Oral Health Intervention.","authors":"Brenda G Grift, Peggy C J M van Spreuwel, Josef J J M Bruers, Katarina Jerković-Ćosić","doi":"10.2147/POR.S493195","DOIUrl":"https://doi.org/10.2147/POR.S493195","url":null,"abstract":"<p><strong>Purpose: </strong>This article examines the implementation of the Toddler Oral Health Intervention (TOHI) protocol, emphasizing feasibility, recruitment, retention, and protocol adherence rather than clinical outcomes. It discusses encountered challenges, necessary adjustments, and key lessons learned during the shift from a traditional randomized controlled trial (RCT) to a pragmatic RCT, thereby offering guidance for future research on similar complex interventions.</p><p><strong>Methods: </strong>The methods of the RCT study protocol are presented, including (1) recruitment of well-baby clinics and oral health coaches, (2) recruitment of the study population; (3) implementation of the intervention and (4) outcome measures, alongside with the process of implementation and gathered feedback needed for required adjustments. A systematic analysis identified key protocol elements, modifications made, and lessons learned.</p><p><strong>Results: </strong>Specific protocol modifications, including adjustments due to the COVID-19 pandemic, demonstrate how a traditional RCT can be adapted to real-world conditions without compromising outcome reliability. Key lessons include the necessity of robust contingency planning to manage unforeseen disruptions, effective recruitment and retention strategies to sustain participant and interventionists engagement, and the adoption of standardized data collection processes to ensure data integrity. These findings underscore the importance of addressing practical and contextual factors alongside measuring effectiveness, ensuring that study outcomes are both applicable and useful in clinical practice.</p><p><strong>Conclusion: </strong>The findings demonstrate that carefully planned and documented pragmatic adjustments not only preserve scientific rigor but also enhance the relevance and applicability of the results, maintaining the methodological robustness expected of traditional RCTs.</p>","PeriodicalId":20399,"journal":{"name":"Pragmatic and Observational Research","volume":"16 ","pages":"111-120"},"PeriodicalIF":2.3,"publicationDate":"2025-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11963812/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143773133","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting Risk of Morbidities Associated with Oral Corticosteroid Prescription for Asthma.","authors":"Brooklyn Stanley, Jatin Chapaneri, Mina Khezrian, Ekaterina Maslova, Soram Patel, Mark Gurnell, Giorgio Walter Canonica, Helen K Reddel, Liam G Heaney, Arnaud Bourdin, David L Neil, Victoria Carter, David B Price","doi":"10.2147/POR.S484146","DOIUrl":"10.2147/POR.S484146","url":null,"abstract":"<p><strong>Background: </strong>Oral corticosteroids (OCS) are commonly used to treat asthma but increase the risks for multiple morbidities; reducing OCS exposure may benefit patients. We analysed independent risk factors and longitudinal changes in OCS usage among patients with asthma to predict future risks of OCS-related adverse outcomes.</p><p><strong>Methods: </strong>Optimum Patient Care Research Database United Kingdom primary care electronic medical records (EMR) from January 1990 to June 2021 were used to select adults (18-93 years) with asthma who had follow-up data from ≥2 years before to ≥3 years after an index visit for active symptoms; this date was defined by the largest pre-visit to post-visit change in mean annual OCS use. OCS usage during every follow-up year was categorised as none, low (mean <2 prescriptions/year), or high (mean ≥2 prescriptions/year). Pre-index to post-index changes between usage categories were calculated. Risk modelling selected cohorts without 17 morbidities (documented pre-index) reported to be associated with OCS exposure, including type 2 diabetes, osteoporosis, hypertension, and pneumonia. Cox regression analyses selected published risk factors associated with each condition and available in EMR for inclusion in proportional hazards models.</p><p><strong>Results: </strong>The pre-index to post-index OCS usage category remained unchanged in 38.6% of patients, increased in 39.2%, and decreased in 22.2%, with 20.7% having no further OCS prescriptions. In models, the risks of all adverse outcomes increased with projected categoric OCS use; for example, hazard ratios for a one-category increment (none to low, low to high) were 1.55 (1.42-1.69) for type 2 diabetes, 1.56 (1.36-1.78) for post-menopausal osteoporosis, 1.05 (1.00-1.10) for hypertension, and 1.67 (1.52-1.83) for pneumonia (all p < 0.001).</p><p><strong>Conclusion: </strong>OCS exposure in this primary care asthma population usually continued longitudinally. Our models predict increased risk of multiple morbidities with higher projected OCS exposure. These findings support early initiation of strategies to minimise OCS use in asthma.</p>","PeriodicalId":20399,"journal":{"name":"Pragmatic and Observational Research","volume":"16 ","pages":"95-109"},"PeriodicalIF":2.3,"publicationDate":"2025-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11954399/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143754303","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Better Understanding Rehabilitation of Motor Symptoms: Insights from the Use of Wearables.","authors":"Yunus Celik, Conor Wall, Jason Moore, Alan Godfrey","doi":"10.2147/POR.S396198","DOIUrl":"10.2147/POR.S396198","url":null,"abstract":"<p><p>Movement disorders present a substantial challenge by adversely affecting daily routines and overall well-being through a diverse spectrum of motor symptoms. Traditionally, motor symptoms have been evaluated through manual observational methods and patient-reported outcomes. While those approaches are valuable, they are limited by their subjectivity. In contrast, wearable technologies (wearables) provide objective assessments while actively supporting rehabilitation through continuous tracking, real-time feedback, and personalized physical therapy-based interventions. The aim of this literature review is to examine current research on the use of wearables in the rehabilitation of motor symptoms, focusing on their features, applications, and impact on improving motor function. By exploring research protocols, metrics, and study findings, this review aims to provide a comprehensive overview of how wearables are being used to support and optimize rehabilitation outcomes. To achieve that aim, a systematic search of the literature was conducted. Findings reveal that gait disturbance and postural balance are the primary motor symptoms extensively studied with tremor and freezing of gait (FoG) also receiving attention. Wearable sensing ranges from bespoke inertial and/or electromyography to commercial units such as personal devices (ie, smartwatch). Interactive (virtual reality, VR and augmented reality, AR) and immersive technologies (headphones), along with wearable robotic systems (exoskeletons), have proven to be effective in improving motor skills. Auditory cueing (via smartwatches or headphones), aids gait training with rhythmic feedback, while visual cues (via VR and AR glasses) enhance balance exercises through real-time feedback. The development of treatment protocols that incorporate personalized cues via wearables could enhance adherence and engagement to potentially lead to long-term improvements. However, evidence on the sustained effectiveness of wearable-based interventions remains limited.</p>","PeriodicalId":20399,"journal":{"name":"Pragmatic and Observational Research","volume":"16 ","pages":"67-93"},"PeriodicalIF":2.3,"publicationDate":"2025-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11930022/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143692998","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Real-World Biologic Use Patterns in Severe Asthma, 2015-2021: The CLEAR Study.","authors":"Trung N Tran, Stephanie Chen, Benjamin Emmanuel, Alan Altraja, Arnaud Bourdin, Chau-Chyun Sheu, Ming-Ju Tsai, Flavia C L Hoyte, Anna Quinton, Bill Cook, Lakmini Bulathsinhala, William Henley, Celine Yun Yi Goh, Yang Liu, Cono Ariti, Victoria Carter, David B Price","doi":"10.2147/POR.S497033","DOIUrl":"10.2147/POR.S497033","url":null,"abstract":"<p><strong>Background: </strong>Biologics targeting immunoglobulin E, interleukin (IL)-4/IL-13 or IL-5 signaling are effective at treating severe asthma; however, individual patients' responses may be suboptimal, leading to therapy switching or stopping. The CLEAR study aimed to assess real-world biologic use patterns and associated clinical outcomes in patients receiving care for severe asthma.</p><p><strong>Methods: </strong>CLEAR was a multicenter, observational study that included adults (≥18 years old) from 23 countries enrolled in the International Severe Asthma Registry between December 2015 and August 2021. Patients who initiated biologic therapy were categorized as continuing the initial biologic for 6 months, switching to another biologic within 6 months or stopping biologic treatment within 6 months. Outcomes were assessed using the closest available data to 12 months after biologic initiation, using propensity score-weighted multivariable regression models.</p><p><strong>Results: </strong>Among 1,859 patients who initiated biologic therapy, 1,116 (60.0%) continued, 474 (25.5%) switched and 269 (14.5%) stopped treatment. Patients who switched or stopped therapy had a higher annualized asthma exacerbation rate post-initiation than those who continued (adjusted incidence rate ratio [aIRR] [95% confidence interval]: switched, 1.83 [1.51, 2.22]; stopped, 1.53 [1.19, 1.95]) and were more likely to have uncontrolled asthma at last assessment (adjusted odds ratio: switched, 5.40 [3.12, 9.33]; stopped, 4.02 [2.32, 6.98]). Compared with those who continued therapy, patients who switched had a higher long-term daily oral corticosteroid dose (adjusted β: 3.77 [1.71, 4.37] mg) and higher rates of hospitalizations (aIRR: 2.58 [1.52, 4.37]) and emergency room visits (aIRR: 2.12 [1.39, 3.24]).</p><p><strong>Conclusion: </strong>Switching or stopping biologic therapy was associated with worse clinical outcomes than continuing the initial therapy.</p>","PeriodicalId":20399,"journal":{"name":"Pragmatic and Observational Research","volume":"16 ","pages":"51-66"},"PeriodicalIF":2.3,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11929530/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143693027","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Prospective Observational Study to Evaluate the Actual Use of 1% Pimecrolimus in Chinese Patients with Mild-to-Moderate Atopic Dermatitis Affecting Sensitive Skin Areas.","authors":"Zhiqiang Xie, Jianmin Chang, Lin Ma, Chunping Shen, Li Zhang, Qian An, Hua Wang, Xia Dou, Yue Zheng, Congxiu Ye, Ying Gao","doi":"10.2147/POR.S496970","DOIUrl":"10.2147/POR.S496970","url":null,"abstract":"<p><strong>Introduction: </strong>Atopic dermatitis (AD) is a common inflammatory skin disorder that affects both children and adults, characterized by pruritus and scaly, dry eczematous lesions. This study aimed to gather knowledge on the actual use and effectiveness of Pimecrolimus (PIM) in Chinese patients with mild-to-moderate AD affecting sensitive skin areas in routine clinical practice.</p><p><strong>Methods: </strong>This multicentre and non-interventional study included 130 subjects from China, divided into two age groups (2-12 years and ≥12 years). The primary endpoint was the change in SCORAD index, in AD areas from inclusion to the end of the PIM treatment period.</p><p><strong>Results: </strong>The primary efficacy analysis showed a significant reduction in the SCORAD index from baseline to the end of the PIM treatment period (p < 0.0001). The mean change in SCORAD index (SD) from baseline at visit 2 and visit 3 was -15.4 (10.50) and -18.6 (11.57), respectively. The mean Investigator's Global Assessment (IGA) score decreased from 2.3 at baseline to 0.7 at the end of the study, while the mean itching score decreased from 4.7 at baseline to 0.9 at the end of the study. The mean duration of PIM use was 38.3 days, with similar durations for patients above and below 12 years. There was progressive improvement in the quality of life of the patients with PIM treatment. The median time to first flare was 100 days and no adverse drug reactions or significant adverse events were reported during the study.</p><p><strong>Discussion: </strong>This study provides robust real-world evidence that PIM 1% cream is effective for the treatment of mild-to-moderate AD in Chinese patients, particularly in sensitive skin areas and paediatric population. PIM also offers a TCS-sparing approach, making it a valuable option for managing mild-to-moderate AD.</p>","PeriodicalId":20399,"journal":{"name":"Pragmatic and Observational Research","volume":"16 ","pages":"39-49"},"PeriodicalIF":2.3,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11895690/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143606252","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Treatment Patterns, Adverse Events, and Clinical Outcomes with Steroidal Mineralocorticoid Receptor Antagonists: A Retrospective Analysis of Administrative Claims Data (RELICS).","authors":"Emma L Richard, Nihar R Desai, Vincent J Willey, Alain Gay, Charlie Scott, Kerstin Folkerts, Elena Pessina, Rakesh Singh, Chia-Chen Teng, Nikolaus G Oberprieler","doi":"10.2147/POR.S489791","DOIUrl":"10.2147/POR.S489791","url":null,"abstract":"<p><strong>Purpose: </strong>This study aimed to describe the characteristics, treatment patterns, adverse events (AEs), and clinical outcomes of patients starting steroidal mineralocorticoid receptor antagonists (sMRAs) in real-world settings.</p><p><strong>Methods: </strong>The RELICS study, complementing the survey-based RELICS-PS study, was a retrospective cohort study conducted using the Healthcare Integrated Research Database (HIRD^®), a single-payer healthcare database with medical and pharmacy claims from health insurance plans across the United States. A cohort of adults initiating sMRAs from January 2016 to June 2021 was divided into six subgroups: three mutually exclusive heart failure (HF) subgroups, two mutually exclusive chronic kidney disease (CKD) subgroups, and \"all other patients\" subgroup, which included those without documented HF or CKD. Outcomes assessed from the first sMRA fill until death, health-plan disenrollment, or June 2022 (whichever came first) included analysis of treatment patterns, AEs, and clinical outcomes. Factors associated with sMRA discontinuation were evaluated with multivariate logistic regression.</p><p><strong>Results: </strong>Of the 224,100 sMRA initiators identified, 76.4% did not have documented HF or CKD (ie, \"all other patients\" subgroup). This subgroup was younger and primarily female. Across all initiators, 72.3% were nonadherent, and 73.0% discontinued treatment within a median of 90 days of initiation. Of these discontinuers, 44.2% restarted treatment within a median of 91 days of discontinuation. Factors decreasing odds of discontinuation across most subgroups included a higher comorbidity burden, use of other cardiovascular medications, and cardiologist prescribing. These findings were consistent across subgroups. AEs and clinical outcomes varied across subgroups in line with baseline comorbidity profiles. Patients with a higher comorbidity burden, such as those with both CKD and T2D rather than CKD alone, experienced worse outcomes.</p><p><strong>Conclusion: </strong>High rates of treatment discontinuation and subsequent restart were observed across all subgroups, implying fluctuating sMRA use. However, heightened cardiovascular risk may decrease the odds of discontinuation.</p>","PeriodicalId":20399,"journal":{"name":"Pragmatic and Observational Research","volume":"16 ","pages":"27-37"},"PeriodicalIF":2.3,"publicationDate":"2025-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11873017/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143543096","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Validation of Mortality Data Sources Compared to the National Death Index in the Healthcare Integrated Research Database.","authors":"Aziza Jamal-Allial, Todd Sponholtz, Shiva K Vojjala, Mark Paullin, Anahit Papazian, Biruk Eshete, Seyed Hamidreza Mahmoudpour, Patrice Verpillat, Daniel C Beachler","doi":"10.2147/POR.S498221","DOIUrl":"10.2147/POR.S498221","url":null,"abstract":"<p><strong>Background: </strong>The National Death Index (NDI) is the gold standard for mortality data in the United States (US) but has a time lag and can be operationally intensive. This validation study assesses the accuracy of various mortality data sources with the NDI.</p><p><strong>Methods: </strong>This validation study is a secondary analysis of an advanced cancer cohort in the US between January 2010 and December 2018, with an established NDI linkage. Mortality data sources, inpatient discharge, disenrollment, death master file (DMF), Center for Medicare and Medicaid Services (CMS), Utilization management data (U.M.), and online obituary data were compared to NDI.</p><p><strong>Results: </strong>Among 40,692 patients, 25,761 (63.3%) had a death date using NDI; the composite algorithm had a sensitivity of 88.9% (95% CI = 88.5%, 89.3%), specificity was 89.1% (95% CI = 88.6%, 89.6%). At the same time, positive predictive value (PPV) was 93.4% (95% CI = 93.1%, 93.7%), negative predictive value (NPV) was 82.3% (95% CI = 81.7%, 82.9%), and when comparing each individual source, each had a high PPV but limited sensitivity.</p><p><strong>Conclusion: </strong>The composite algorithm was demonstrated to be a sensitive and precise measure of mortality, while individual database sources were accurate but had limited sensitivity.</p>","PeriodicalId":20399,"journal":{"name":"Pragmatic and Observational Research","volume":"16 ","pages":"19-25"},"PeriodicalIF":2.3,"publicationDate":"2025-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11812554/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143399796","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Review of Challenges in Performing Real-World Evidence Studies for Nonprescription Products.","authors":"Christian Maihöfner, Theresa Mallick-Searle, Jan Vollert, Pranab Kalita, Vidhu Sood Sethi","doi":"10.2147/POR.S504709","DOIUrl":"10.2147/POR.S504709","url":null,"abstract":"<p><p>In recent years, regulatory authorities have signaled a willingness to consider real-world evidence (RWE) data to support applications for new claims and indications for pharmaceuticals. Historically, RWE studies have been the domain of prescription drugs, driven by the fact that clinical data on patients are routinely captured in medical records, claims databases, registries, etc. However, RWE reports of nonprescription drugs and supplements are relatively sparse due to methodological gaps in this area. The objective of this narrative review is to identify which RWE methodologies have been used to study nonprescription products. A total of 49 articles were included based on literature searches. Label comprehension studies, used to support prescription-to-nonprescription switches, are useful in determining how nonprescription products will be used; however, they provide no actual clinical data. The most common RWE studies of nonprescription products were cross-sectional surveys, which investigated a broad range of indications and were conducted in an array of settings, including online, by phone, point-of-sale (pharmacy), outpatient clinics, and shopping malls. However, while this type of study is effective for identifying use patterns and attitudes in the general population, recall bias limits the ability to collect safety and effectiveness data. Studies of electronic medical records and claims databases are hampered by incomplete or absent capturing of data on nonprescription products. As a result, most RWE studies to date have provided limited useful information. Although case reports and expert opinion should not be discounted, in the absence of other information they provide few actual data. Novel approaches using smartphone apps and artificial intelligence may provide new opportunities to collect RWE for nonprescription products, but these areas of research are in their infancy. Overall, there is a need to develop standards for execution of RWE studies of nonprescription products in terms of endpoints, study design, and study quality.</p>","PeriodicalId":20399,"journal":{"name":"Pragmatic and Observational Research","volume":"16 ","pages":"7-18"},"PeriodicalIF":2.3,"publicationDate":"2025-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11771160/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143053317","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Measuring Diagnostic Quality: The Capacity of Routinely Collected Data and Applications to Chronic Respiratory Disease.","authors":"Rachael Mountain, Timothy Gatheral, Patrick Haslam, Kelly Heys, Jo Knight","doi":"10.2147/POR.S430705","DOIUrl":"10.2147/POR.S430705","url":null,"abstract":"","PeriodicalId":20399,"journal":{"name":"Pragmatic and Observational Research","volume":"16 ","pages":"1-6"},"PeriodicalIF":2.3,"publicationDate":"2025-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11745041/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143010407","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Detection of Patient-Level Immunotherapy-Related Adverse Events (irAEs) from Clinical Narratives of Electronic Health Records: A High-Sensitivity Artificial Intelligence Model.","authors":"Md Muntasir Zitu, Margaret E Gatti-Mays, Kai C Johnson, Shijun Zhang, Aditi Shendre, Mohamed I Elsaid, Lang Li","doi":"10.2147/POR.S468253","DOIUrl":"10.2147/POR.S468253","url":null,"abstract":"<p><strong>Purpose: </strong>We developed an artificial intelligence (AI) model to detect immunotherapy -related adverse events (irAEs) from clinical narratives of electronic health records (EHRs) at the patient level.</p><p><strong>Patients and methods: </strong>Training data, used for internal validation of the AI model, comprised 1230 clinical notes from 30 patients at The Ohio State University James Cancer Hospital-20 patients who experienced irAEs and ten who did not. 3256 clinical notes of 50 patients were utilized for external validation of the AI model.</p><p><strong>Results: </strong>Use of a leave-one-out cross-validation technique for internal validation among those 30 patients yielded accurate identification of 19 of 20 with irAEs (positive patients; 95% sensitivity) and correct dissociation of eight of ten without (negative patients; 80% specificity). External validation on 3256 clinical notes of 50 patients yielded high sensitivity (95%) but moderate specificity (64%). If we improve the model's specificity to 100%, it could eliminate the need to manually review 2500 of those 3256 clinical notes (77%).</p><p><strong>Conclusion: </strong>Combined use of this AI model with the manual review of clinical notes will improve both sensitivity and specificity in the detection of irAEs, decreasing workload and costs and facilitating the development of improved immunotherapies.</p>","PeriodicalId":20399,"journal":{"name":"Pragmatic and Observational Research","volume":"15 ","pages":"243-252"},"PeriodicalIF":2.3,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11668329/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142886144","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}