Electronic health records to facilitate clinical research

Leveraging undecided cases in chart-reviewed phenotypes to enhance EHR-based association studies

OBJECTIVES

In electronic health record (EHR)-based association studies, phenotyping algorithms efficiently classify patient clinical outcomes into binary categories but are susceptible to misclassification errors. The gold standard, manual chart review, involves clinicians determining the true disease status based on their assessment of health records. These clinicians-labeled phenotypes are labor-intensive and typically limited to a small subset of patients, potentially introducing a third "undecided" category when phenotypes are indeterminate. We aim to effectively integrate the algorithm-derived and chart-reviewed outcomes when both are available in EHR-based association studies.

MATERIAL AND METHODS

We propose an augmented estimation method that combines the binary algorithm-derived phenotypes for the entire cohort with the trinary chart-reviewed phenotypes for a small, selected subset. Additionally, a cost-effective outcome-dependent sampling strategy is used to address the rare disease scenarios. The proposed trinary chart-reviewed phenotype integrated cost-effective augmented estimation (TriCA) was evaluated across a wide range of simulation settings and real-world applications, including using EHR data on Alzheimer's disease and related dementias (ADRD) from the OneFlorida + Clinical Research Network, and using cohort data on second breast cancer events (SBCE) from the Kaiser Permanente Washington.

RESULTS

Compared to estimation based on random sampling, our augmented method improved mean square error by up to 28.3% in simulation studies; compared to estimation using only trinary chart-reviewed phenotypes, our method improved efficiency by up to 33.3% in ADRD data and 50.8% in SBCE data.

DISCUSSION

Our simulation studies and real-world applications demonstrate that, compared to existing methods, the proposed method provides unbiased estimates with higher statistical efficiency.

CONCLUSION

The proposed method effectively combined binary algorithm-derived phenotypes for the whole cohort with trinary chart-reviewed outcomes for a limited validation set, making it applicable to a broader range of applications and enhancing risk factor identification in EHR-based association studies.

Machine-Learned Codes from EHR Data Predict Hard Outcomes Better than Human-Assigned ICD Codes

We used machine learning (ML) to characterize 894,154 medical records of outpatient visits from the Veterans Administration Central Data Warehouse (VA CDW) by the likelihood of assignment of 200 International Classification of Diseases (ICD) code blocks. Using four different predictive models, we found the ML-derived predictions for the code blocks were consistently more effective in predicting death or 90-day rehospitalization than the assigned code block in the record. We reviewed records of ICD chapter assignments. The review revealed that the ML-predicted chapter assignments were consistently better than those humanly assigned. Impact factor analysis, a method of explanation of AI findings that was developed in our group, demonstrated little effect on any one assigned ICD code block but a marked impact on the ML-derived code blocks of kidney disease as well as several other morbidities. In this study, machine learning was much better than human code assignment at predicting the relatively rare outcomes of death or rehospitalization. Future work will address generalizability using other datasets, as well as addressing coding that is more nuanced than that of the categorization provided by code blocks.

Validation of an Algorithm to Identify End-Stage Kidney Disease in Electronic Health Records Data

RATIONALE & OBJECTIVES

Accurate ascertainment of end-stage kidney disease (ESKD) in electronic health records (EHRs) data is important for much epidemiological research. This study aimed to develop and validate an algorithm using diagnosis and procedure codes to identify patients with ESKD (treated with maintenance dialysis or kidney transplantation) in EHRs data.

STUDY DESIGN

Study of diagnostic algorithms.

SETTING & PARTICIPANTS

The development cohort included 559,615 patients treated at the Geisinger Health System (January 1996-June 2018). The validation cohort included 767,186 patients treated at New York University Langone Health System (January 2018-December 2020).

ALGORITHMS COMPARED

An algorithm developed using diagnosis and procedure codes compared to a nominal gold standard designation within the United States Renal Data System (USRDS) data. The performance of the algorithm was characterized by sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). The dates of incident ESKD between the algorithm and USRDS were compared in a subset of cases.

OUTCOMES

ESKD (maintenance dialysis, prior recipient of a kidney transplant, or kidney transplantation surgery) cases.

RESULTS

In Geisinger, we developed the ESKD algorithm that identified 4,766 (0.85%) ESKD cases, while there were 5,155 (0.92%) ESKD cases reported by the USRDS. The sensitivity, specificity, PPV, and NPV of the algorithm were 73.9% (95% CI, 72.7-75.1%), 99.83% (99.82-99.84%), 79.9% (78.9-81.0%), and 99.76% (99.75-99.77%), respectively. When applying the algorithm to New York University Langone Health System data, the sensitivity, specificity, PPV, and NPV were 71.8% (95% CI, 70.7-73.0%), 99.95% (99.95-99.96%), 91.6% (90.8-92.4%), and 99.79 (99.78-99.80%), respectively. The median (interquartile range) difference between dates of incident ESKD (algorithms minus USRDS) were -3 (-21 to 83) days in Geisinger and 0 (-12 to 69) days in New York University Langone Health.

LIMITATIONS

Use of structured EHRs data only.

CONCLUSIONS

Algorithms combining diagnosis and procedure codes show high specificity and modest sensitivity for identifying patients with ESKD, providing a research tool to inform future EHR-based studies.

Longitudinal Masked Representation Learning for Pulmonary Nodule Diagnosis from Language Embedded EHRs

Electronic health records (EHRs) are a rich source of clinical data, yet exploiting longitudinal signals for pulmonary nodule diagnosis remains challenging due to the administrative noise and high level of clinical abstraction present in these records. Because of this complexity, classification models are prone to overfitting when labeled data is scarce. This study explores masked representation learning (MRL) as a strategy to improve pulmonary nodule diagnosis by modeling longitudinal EHRs across multiple modalities: clinical conditions, procedures, and medications. We leverage a web-scale text embedding model to encode EHR event streams into semantically embedded sequences. We then pretrain a bidirectional transformer using MRL conditioned on time encodings on a large cohort of general pulmonary conditions from our home institution. Evaluation on a cohort of diagnosed pulmonary nodules demonstrates significant improvement in diagnosis accuracy with a model finetuned from MRL (0.781 AUC, 95% CI: [0.780, 0.782]) compared to a supervised model with the same architecture (0.768 AUC, 95% CI: [0.766, 0.770]) when integrating all three modalities. These findings suggest that language-embedded MRL can facilitate downstream clinical classification, offering potential advancements in the comprehensive analysis of longitudinal EHR modalities.

Subtypes and service utilization among opioid use disorder patients at a community health center: findings from a medically underserved urban area of the Northeastern United States

BACKGROUND

Opioid use disorder often co-occurs with other mental health and substance use disorders. Identifying clusters of individuals receiving treatment for opioid use disorder based on co-diagnosed conditions, healthcare plans, and service utilization over a seven-year treatment period provides insight into service needs. Objectives included [1] characterizing the sample [2], examining subtypes of the sample using cluster analysis, and [3] identifying differences in Current Procedural Terminology by subtype to examine service utilization among identified clusters.

METHODS

This study uses secondary data from the electronic medical records of a community health center in a large urban area in the Northeastern United States from 2015 to 2021. The study sample included N = 705 adults who had an opioid use disorder diagnosis as indicated by the community health center's electronic medical records. Measures include [1] age [2], race and ethnicity [3], sex [4], healthcare plan(s) [5], co-occurring mental health disorder [6], co-occurring substance use disorder [7], co-occurring mental health disorder or substance use disorder, and [8] Current Procedural Terminology codes for behavioral health service utilization. Cluster analysis was used to examine the sample. These clusters were then analyzed for service utilization with a one-way analysis of variance.

RESULTS

The cluster analysis identified six clusters with an average silhouette of 0.5, indicating good clustering. These six clusters were operationalized as [1] Medicare/Medicaid healthcare plan with substance use disorder needs [2], Private pay and charity care healthcare plan with cocaine use disorder needs [3], Medicare/Medicaid and other publicly-funded healthcare plans with mood disorder needs [4], Private healthcare plan with low co-occurring disorder needs [5], Other publicly-funded healthcare plan with cannabis use disorder needs [6], Medicare/Medicaid healthcare plan with mental health disorder needs. Service utilization differed between these clusters with cluster mean differences for psychotherapy sessions (F = 8.55, p < 0.001), psychiatric sessions (F = 22.72, p < 0.001), and group therapy sessions (F = 10.76, p < 0.001).

CONCLUSIONS

This study highlights the importance of comprehensive and integrated treatment for substance use disorders and mental health disorders, particularly for those in underserved communities. Healthcare coverage, a socioeconomic factor that impacts access to care, is critical in distinguishing treatment needs and utilization.

Advancements in Electronic Medical Records for Clinical Trials: Enhancing Data Management and Research Efficiency

Recent advancements in electronic medical records (EMRs) have transformed clinical trials and healthcare systems by improving data accuracy, regulatory compliance, and integration with decision support tools. These innovations enhance trial efficiency, streamline patient recruitment, and enable large-scale data analysis while bridging clinical practice with research. Despite these benefits, challenges such as data standardization, privacy concerns, and usability issues persist. Overcoming these barriers through policy reforms, technological innovations, and robust methodologies is essential to maximizing the potential of EMRs. We examine current developments, challenges, and future directions for optimizing EMRs in clinical trials and healthcare delivery.

Advancing Precision Oncology Through Modeling of Longitudinal and Multimodal Data

Cancer evolves continuously over time through a complex interplay of genetic, epigenetic, microenvironmental, and phenotypic changes. This dynamic behavior drives uncontrolled cell growth, metastasis, immune evasion, and therapy resistance, posing challenges for effective monitoring and treatment. However, today's data-driven research in oncology has primarily focused on cross-sectional analysis using data from a single modality, limiting the ability to fully characterize and interpret the disease's dynamic heterogeneity. Advances in multiscale data collection and computational methods now enable the discovery of longitudinal multimodal biomarkers for precision oncology. Longitudinal data reveal patterns of disease progression and treatment response that are not evident from single-timepoint data, enabling timely abnormality detection and dynamic treatment adaptation. Multimodal data integration offers complementary information from diverse sources for more precise risk assessment and targeting of cancer therapy. In this review, we survey methods of longitudinal and multimodal modeling, highlighting their synergy in providing multifaceted insights for personalized care tailored to the unique characteristics of a patient's cancer. We summarize the current challenges and future directions of longitudinal multimodal analysis in advancing precision oncology.

Bridging Data Silos in Oncology with Modular Software for Federated Analysis on Fast Healthcare Interoperability Resources: Multisite Implementation Study

BACKGROUND

Real-world data (RWD) from sources like administrative claims, electronic health records, and cancer registries offer insights into patient populations beyond the tightly regulated environment of randomized controlled trials. To leverage this and to advance cancer research, 6 university hospitals in Bavaria have established a joint research IT infrastructure.

OBJECTIVE

This study aimed to outline the design, implementation, and deployment of a modular data transformation pipeline that transforms oncological RWD into a Health Level 7 (HL7) Fast Healthcare Interoperability Resources (FHIR) format and then into a tabular format in preparation for a federated analysis (FA) across the 6 Bavarian Cancer Research Center university hospitals.

METHODS

To harness RWD effectively, we designed a pipeline to convert the oncological basic dataset (oBDS) into HL7 FHIR format and prepare it for FA. The pipeline handles diverse IT infrastructures and systems while maintaining privacy by keeping data decentralized for analysis. To assess the functionality and validity of our implementation, we defined a cohort to address two specific medical research questions. We evaluated our findings by comparing the results of the FA with reports from the Bavarian Cancer Registry and the original data from local tumor documentation systems.

RESULTS

We conducted an FA of 17,885 cancer cases from 2021/2022. Breast cancer was the most common diagnosis at 3 sites, prostate cancer ranked in the top 2 at 4 sites, and malignant melanoma was notably prevalent. Gender-specific trends showed larynx and esophagus cancers were more common in males, while breast and thyroid cancers were more frequent in females. Discrepancies between the Bavarian Cancer Registry and our data, such as higher rates of malignant melanoma (3400/63,771, 5.3% vs 1921/17,885, 10.7%) and lower representation of colorectal cancers (8100/63,771, 12.7% vs 1187/17,885, 6.6%) likely result from differences in the time periods analyzed (2019 vs 2021/2022) and the scope of data sources used. The Bavarian Cancer Registry reports approximately 3 times more cancer cases than the 6 university hospitals alone.

CONCLUSIONS

The modular pipeline successfully transformed oncological RWD across 6 hospitals, and the federated approach preserved privacy while enabling comprehensive analysis. Future work will add support for recent oBDS versions, automate data quality checks, and integrate additional clinical data. Our findings highlight the potential of federated health data networks and lay the groundwork for future research that can leverage high-quality RWD, aiming to contribute valuable knowledge to the field of cancer research.

Advancing birth cohort studies using administrative and other research-independent data repositories: Opportunities and challenges

The birth cohort study design is an essential epidemiologic tool for investigating the developmental origins of health and disease. Birth cohorts have greatly improved the etiologic understanding of asthma and allergic diseases, setting the stage for advancements in translational interventions. Increasingly, investigators leverage data repositories that have been compiled and maintained independently of research investigations (administrative data) to establish large birth cohorts or to augment data generated through active participant interaction. In many cases, administrative data can greatly enhance the capacity of birth cohorts to achieve their scientific goals. However, investigators must be wary of common pitfalls and carefully consider whether administrative data are well suited to the scientific inquiry. This article reviews the strengths and challenges of using administrative data and the pragmatic solutions that have been developed to optimize their use in birth cohorts. As birth cohorts continue to play an important role in understanding the etiology of early-life disease, unleashing the power of administrative data will greatly assist in this scientific process.

The Case for Aggregated Rehabilitation-Relevant Data Across Health Care Systems and Settings

Health care value, quantified as outcome per unit cost, requires knowing which outcomes are influenced by which intervention at what cost. The value of rehabilitation is still largely unknown. Much of the reason for this limited evidence is historically poor standardization and collection of rehabilitation interventions, and objectively measured outcomes across care settings, care providers, and health care systems. The purposeful standardization and aggregation of rehabilitation-relevant data about interventions, cost, and outcomes from routine clinical practices offers potential to understand and improve the value of rehabilitation. This perspective details the critical need for rehabilitation-relevant data that are aggregated across settings, providers, and systems and proposes 3 options to meet this need, including (1) integrating rehabilitation-relevant data into existing research registry databases that are condition specific, (2) adding rehabilitation-relevant data to federally funded research networks, and (3) creating a novel rehabilitation registry database. There must be continued pursuit of discovering which rehabilitation interventions achieve which specific outcomes, in which settings, for which patients, and at what costs. Successfully aggregating rehabilitation-relevant data is critical for generating evidence that answers these key questions about the value of rehabilitation.

Social Vulnerability Index and Dental Caries in Children: An Exploratory Study

OBJECTIVE

This retrospective cross-sectional study evaluated the association between caries outcomes in a pediatric population visiting a dental clinic and the social vulnerability index, an area-based measure capturing 4 main social determinants of health: socioeconomic status, household composition/disability, minority status/language, and housing/transportation.

METHODS

The Centers for Disease Control Social Vulnerability Index (SVI) and electronic dental record data of children (0 to 18 y) reporting a caries diagnosis at the Children's Hospital Colorado in 2020 were extracted for 9,201 children. Logistic regressions were used to test the association between SVI and the presence or absence of dental caries, adjusting for age, sex, ethnicity, and race.

RESULTS

Children with a caries diagnosis had a greater mean overall SVI percentile (62.0, standard deviation [SD] = 29.1) compared with patients without a caries diagnosis (59.1, SD = 29.8; P < 0.001). With each 10-point increase in the overall SVI percentile, having a caries diagnosis visit was 2.7% more likely compared with having a visit without a caries diagnosis (odds ratio [OR] 1.027, 95% confidence interval [CI] 1.012, 1.042; P = 0.0004). Those with an overall SVI percentile between 51 and 75 were 23% more likely to have a caries diagnosis compared with those with a percentile ≤25 (OR 1.23, 95% CI 1.07, 1.42; P = 0.003), and those with a percentile >75 were 23.6% more likely to have a caries diagnosis compared with those with a percentile ≤25 (OR 1.236, 95% CI 1.09, 1.40; P = 0.001).

CONCLUSION

Children (0 to 18 y) living in socially vulnerable environments or areas were more likely to have a caries diagnosis at their dental exam.Knowledge Transfer Statement:This study showed an association between social determinants of health demonstrating social vulnerability and dental caries in children. Ultimately, understanding upstream factors for children living in socially vulnerable areas could support policymakers in creating more effective policies to support socially vulnerable populations.

Machine learning in cardiovascular risk assessment: Towards a precision medicine approach

Cardiovascular diseases remain the leading cause of global morbidity and mortality. Validated risk scores are the basis of guideline-recommended care, but most scores lack the capacity to integrate complex and multidimensional data. Limitations inherent to traditional risk prediction models and the growing burden of residual cardiovascular risk highlight the need for refined strategies that go beyond conventional paradigms. Artificial intelligence and machine learning (ML) provide unique opportunities to refine cardiovascular risk assessment and surveillance through the integration of diverse data types and sources, including clinical, electrocardiographic, imaging and multi-omics derived data. In fact, ML models, such as deep neural networks, can handle high-dimensional data through which phenotyping and cardiovascular risk assessment across diverse patient populations become much more precise, fostering a paradigm shift towards more personalized care. Here, we review the role of ML in advancing cardiovascular risk assessment and discuss its potential to identify novel therapeutic targets and to improve prevention strategies. We also discuss key challenges inherent to ML, such as data quality, standardized reporting, model transparency and validation, and discuss barriers in its clinical translation. We highlight the transformative potential of ML in precision cardiology and advocate for more personalized cardiovascular prevention strategies that go beyond previous notions.

Harnessing Electronic Health Records and Artificial Intelligence for Enhanced Cardiovascular Risk Prediction: A Comprehensive Review

Electronic health records (EHR) have revolutionized cardiovascular disease (CVD) research by enabling comprehensive, large-scale, and dynamic data collection. Integrating EHR data with advanced analytical methods, including artificial intelligence (AI), transforms CVD risk prediction and management methodologies. This review examines the advancements and challenges of using EHR in developing CVD prediction models, covering traditional and AI-based approaches. While EHR-based CVD risk prediction has greatly improved, moving from models that integrate real-world data on medication use and imaging, challenges persist regarding data quality, standardization across health care systems, and geographic variability. The complexity of EHR data requires sophisticated computational methods and multidisciplinary approaches for effective CVD risk modeling. AI's deep learning enhances prediction performance but faces limitations in interpretability and the need for validation and recalibration for diverse populations. The future of CVD risk prediction and management increasingly depends on using EHR and AI technologies effectively. Addressing data quality issues and overcoming limitations from retrospective data analysis are critical for improving the reliability and applicability of risk prediction models. Integrating multidimensional data, including environmental, lifestyle, social, and genomic factors, could significantly enhance risk assessment. These models require continuous validation and recalibration to ensure their adaptability to diverse populations and evolving health care environments, providing reassurance about their reliability.

Identifying patients with neurofibromatosis type 1 related optic pathway glioma using the OMOP CDM

Neurofibromatosis type 1 (NF1) is a rare tumour predisposition syndrome. Optic pathway gliomas (NF1-related OPG) are a well-characterised tumour type. There is great need for tools that can efficiently identify patients with NF1-related OPG at hospitals. Computable phenotypes algorithms can be used to find patients with certain clinical features in an electronic database. We developed computable phenotype algorithms using the Observational Medical Outcome Partnership (OMOP) Common Data Model. We subsequently assessed if these algorithms could identify patients with NF1-related OPG in an electronic health records (EHR) derived database. We created phenotype algorithms based on diagnosis codes, visits, and radiologic procedures. These phenotypes were applied to the EHR-derived database of an academic hospital. To assess the performance of the phenotypes, we calculated the precision, recall, and F2 score against a list of known cases (n = 61), provided by a clinician. To evaluate the ability of the phenotypes to identify additional cases, we manually reviewed the predicted positives of each phenotype algorithm. The phenotype algorithm based on the diagnosis codes 'Neurofibromatosis syndrome' and 'Neoplasm of optic nerve' performed best (precision = 1.000, recall = 0.614, F2-score = 0.665). The phenotype 'Neurofibromatosis syndrome and three or more Ophthalmology visits and one or more MRI of brain' performed best of the phenotypes based on visits and radiologic procedures (precision = 0.489, recall = 0.511, F2-score = 0.507). Generally, increased precision came at the cost of a decrease in recall. Following review of the predicted positives of each phenotype, 27 additional cases were identified. OMOP computable phenotype algorithms successfully identified NF1-related OPG patients in an EHR-derived database. They provided swift insight into the number of NF1-related OPG cases and were able to identify additional cases, which were not included in the original list of known cases. Phenotype algorithms created with OMOP could be an invaluable tool to facilitate patient screening, especially in multi-centric trials for rare diseases.

RAMIE: retrieval-augmented multi-task information extraction with large language models on dietary supplements

OBJECTIVE

To develop an advanced multi-task large language model (LLM) framework for extracting diverse types of information about dietary supplements (DSs) from clinical records.

METHODS

We focused on 4 core DS information extraction tasks: named entity recognition (2 949 clinical sentences), relation extraction (4 892 sentences), triple extraction (2 949 sentences), and usage classification (2 460 sentences). To address these tasks, we introduced the retrieval-augmented multi-task information extraction (RAMIE) framework, which incorporates: (1) instruction fine-tuning with task-specific prompts; (2) multi-task training of LLMs to enhance storage efficiency and reduce training costs; and (3) retrieval-augmented generation, which retrieves similar examples from the training set to improve task performance. We compared the performance of RAMIE to LLMs with instruction fine-tuning alone and conducted an ablation study to evaluate the individual contributions of multi-task learning and retrieval-augmented generation to overall performance improvements.

RESULTS

Using the RAMIE framework, Llama2-13B achieved an F1 score of 87.39 on the named entity recognition task, reflecting a 3.51% improvement. It also excelled in the relation extraction task with an F1 score of 93.74, a 1.15% improvement. For the triple extraction task, Llama2-7B achieved an F1 score of 79.45, representing a significant 14.26% improvement. MedAlpaca-7B delivered the highest F1 score of 93.45 on the usage classification task, with a 0.94% improvement. The ablation study highlighted that while multi-task learning improved efficiency with a minor trade-off in performance, the inclusion of retrieval-augmented generation significantly enhanced overall accuracy across tasks.

CONCLUSION

The RAMIE framework demonstrates substantial improvements in multi-task information extraction for DS-related data from clinical records.

Healthcare trajectories of aging individuals during their last year of life: application of process mining methods to administrative health databases

CONTEXT

World is aging and the prevalence of chronic diseases is raising with age, increasing financial strain on organizations but also affecting patients' quality of life until death. Research on healthcare trajectories has gained importance, as it can help anticipate patients' needs and optimize service organization. In an overburdened system, it is essential to develop automated methods based on comprehensive and reliable and already available data to model and predict healthcare trajectories and future utilization. Process mining, a family of process management and data science techniques used to derive insights from the data generated by a process, can be a solid candidate to provide a useful tool to support decision-making.

OBJECTIVE

We aimed to (1) identify the healthcare baseline trajectories during the last year of life, (2) identify the differences in trajectories according to medical condition, and (3) identify adequate settings to provide a useful output.

METHODS

We applied process mining techniques on a retrospective longitudinal cohort of 21,255 individuals who died between April 1, 2014, and March 31, 2018, and were at least 66 years or older at death. We used 6 different administrative health databases (emergency visit, hospitalisation, homecare, medical consultation, death register and administrative), to model individuals' healthcare trajectories during their last year of life.

RESULTS

Three main trajectories of healthcare utilization were highlighted: (i) mainly accommodating a long-term care center; (ii) services provided by local community centers in combination with a high proportion of medical consultations and acute care (emergency and hospital); and (iii) combination of consultations, emergency visits and hospitalization with no other management by local community centers or LTCs. Stratifying according to the cause of death highlighted that LTC accommodation was preponderant for individuals who died of physical and cognitive frailty. Conversely, services offered by local community centers were more prevalent among individuals who died of a terminal illness. This difference is potentially related to the access to and use of palliative care at the end-of-life, especially home palliative care implementation.

CONCLUSION

Despite some limitations related to data and visual limitations, process mining seems to be a method that is both relevant and simple to implement. It provides a visual representation of the processes recorded in various health system databases and allows for the visualization of the different trajectories of healthcare utilization.

Effectiveness of electronic medical record-based strategies for death and hospital admission endpoint capture in pragmatic clinical trials

OBJECTIVE

Event capture in clinical trials is resource-intensive, and electronic medical records (EMRs) offer a potential solution. This study develops algorithms for EMR-based death and hospitalization capture and compares them with traditional event capture methods.

MATERIALS AND METHODS

We compared the effectiveness of EMR-based event capture and site-captured events adjudicated by a clinical endpoint committee in the multi-center INfluenza Vaccine to Effectively Stop cardio Thoracic Events and Decompensated heart failure (INVESTED) trial for participants from the Veterans Affairs healthcare system. Varying time windows around event dates were used to optimize events matching. The algorithms were externally validated for heart failure hospitalizations in the Medical Information Mart for Intensive Care (MIMIC)-IV database.

RESULTS

We observed 100% sensitivity for death events with a 1-day window. Sensitivity for cardiovascular, heart failure, pulmonary, and nonspecific cardiopulmonary hospitalizations using discharge diagnosis codes varied between 75% and 95%. Including Centers for Medicare & Medicaid Services data improved sensitivity with no meaningful decrease in specificity. The MIMIC-IV analysis showed 82% sensitivity and 99% specificity for heart failure hospitalizations.

DISCUSSION

EMR-based method accurately identifies all-cause mortality and demonstrates high accuracy for cardiopulmonary hospitalizations. This study underscores the importance of optimal time windows, data completeness, and domain variability in EMR systems.

CONCLUSION

EMR-based methods are effective strategies for capturing death and hospitalizations in clinical trials; however, their effectiveness may be influenced by the complexity of events and domain variability across different EMR systems. Nonetheless, EMR-based methods can serve as a valuable complement to traditional methods.

Digital health in cardiovascular medicine: An overview of key applications and clinical impact by the Portuguese Society of Cardiology Study Group on Digital Health

Digital health interventions including telehealth, mobile health, artificial intelligence, big data, robotics, extended reality, computational and high-fidelity bench simulations are an integral part of the path toward precision medicine. Current applications encompass risk factor modification, chronic disease management, clinical decision support, diagnostics interpretation, preprocedural planning, evidence generation, education, and training. Despite the acknowledged potential, their development and implementation have faced several challenges and constraints, meaning few digital health tools have reached daily clinical practice. As a result, the Portuguese Society of Cardiology Study Group on Digital Health set out to outline the main digital health applications, address some of the roadblocks hampering large-scale deployment, and discuss future directions in support of cardiovascular health at large.

Association of Polypharmacy and Burden of Comorbidities on COVID-19 Adverse Outcomes in People with Type 1 or Type 2 Diabetes

INTRODUCTION

It is widely accepted that the higher the number of medications prescribed and taken by an individual, the higher the risk of poor health outcomes. We have investigated whether polypharmacy and comorbidities conveyed more risk of adverse health outcomes following COVID-19 infection (as a paradigm of serious viral infections in general) in people with type 1 diabetes (T1DM) or type 2 diabetes (T2DM).

METHODS

The Greater Manchester Care Record (GMCR) is an integrated database of electronic health records containing data collected from 433 general practices in Greater Manchester. Baseline demographic information (age, body mass index [BMI], gender, ethnicity, smoking status, deprivation index), hospital admission or death within 28 days of infection were extracted for adults (18+) diagnosed with either T1DM or T2DM.

RESULTS

The study cohort included individuals diagnosed as T1DM and T2DM separately. Across the Greater Manchester Region, a total of 145,907 individuals were diagnosed with T2DM and 9705 were diagnosed with T1DM. For the T2DM individuals, 45.2% were women and for the T1DM individuals, 42.7% were women. For T2DM, 16-20 medications (p = 0.005; odds ratio [OR] [95% confidence interval (CI) 2.375 [1.306-4.319]) and > 20 medications (p < 0.001; OR [95% CI] 3.141 [1.755-5.621]) were associated with increased risk of death following COVID-19 infection. Increased risk of hospital admissions in T2DM individuals was associated with 11 to 15 medications (p = 0.013; OR = 1.341 (95% CI) [1.063-1.692]). This was independent of comorbidities, metabolic and demographic factors. For T1DM, there was no association of polypharmacy with hospital admission. Additionally, respiratory, cardiovascular/cerebrovascular and gastrointestinal conditions were associated with increased risk of hospital admissions and deaths in T2DM (p < 0.001). Many comorbidities were common across both T1DM and T2DM.

CONCLUSIONS

We have shown in T2DM an independent association of multiple medications taken from 11 upwards with adverse health consequences following COVID-19 infection. We also found that individuals with diabetes develop comorbidities that were common across both T1DM and T2DM. This study has laid the foundation for future investigations into the way that complex pharmacological interactions may influence clinical outcomes in people with T2DM.

Evaluating the quality of prostate cancer diagnosis recording in CPRD GOLD and CPRD Aurum primary care databases for observational research: A study using linked English electronic health records

BACKGROUND

Primary care data in the UK are widely used for cancer research, but the reliability of recording key events like diagnoses remains uncertain. Although data linkage can improve reliability, its costs, time requirements, and sample size constraints may discourage its use. We evaluated accuracy, completeness, and date concordance of prostate cancer (PCa) diagnosis recording in Clinical Practice Research Datalink (CPRD) GOLD and Aurum compared to linked Cancer Registry (CR) and Hospital Episode Statistics (HES) Admitted Patient Care (APC) in England.

METHODS

Incident PCa diagnoses (2000-2016) for males aged ≥46 at diagnosis who remained registered with their General Practitioner (GP) by age 65 and were recorded in at least one data source were analysed. Accuracy was the proportion of diagnoses recorded in GOLD or Aurum with a corresponding record in CR or HES. Completeness was the proportion of CR or HES diagnoses with a corresponding record in GOLD or Aurum.

RESULTS

The final cohorts for comparisons included 29,500 records for GOLD and 26,475 for Aurum. Compared to CR, GOLD was 86 % accurate and 65 % complete, while Aurum was 87 % accurate and 77 % complete. Compared to HES, GOLD was 76 % accurate and 60 % complete, and Aurum was 79 % accurate and 70 % complete. Concordance in diagnosis dates improved over time in both GOLD and Aurum, with 93 % of diagnoses recorded within a year compared to CR, and 66 % (GOLD) and 71 % (Aurum) compared to HES. Delays of 2-3 weeks in primary care diagnosis recording were observed compared to CR, whereas most diagnoses appeared at least 3 months earlier in primary care than in HES.

CONCLUSIONS

Aurum demonstrated better accuracy and completeness for PCa diagnosis recording than GOLD. However, linkage to HES or CR is recommended for improved case capture. Researchers should address the limitations of each data source to ensure research validity.

Contemporary Burden of Cardiovascular Disease in Pregnancy: Insights from a Real-World Pregnancy Electronic Health Record Cohort

Importance

Cardiovascular disease (CVD) is the leading cause of maternal morbidity and mortality, however the contemporary burden and secular trends in pregnancy-related CV complications are not well characterized.

Objective

We sought to examine contemporary trends in prevalence of maternal cardiometabolic comorbidities and established CVD, as well as future pregnancy-related CV complications across a large multi-institutional health system.

Design

Retrospective analysis of longitudinal electronic health record (EHR)-based cohort of pregnancies.

Setting

Multi-institutional healthcare network in New England.

Participants

Pregnancy encounters between 2001 to 2019 identified using diagnosis and procedure codes followed by manual adjudication within a previously validated primary care EHR cohort. Estimated gestational ages recovered from unstructured notes using regular expressions (RegEx) were used to define individual pregnancy episodes.

Main Outcomes and Measures

We quantified the prevalence of maternal cardiometabolic comorbidities and established CVD at time of pregnancy, as well as the incidence of pregnancy-related CV complications assessed within 1 year postpartum. We examined trends in cardiometabolic risk factors and CVD burden over nearly two decades.

Results

Our EHR pregnancy cohort comprised 57,683 pregnancies among 38,997 individuals (mean age range at start of pregnancy 27 to 37 years). RegEx recovered gestational age for 74% of pregnancies, with good correlation between gestational age ascertained via RegEx vs manual review (Pearson r 0.9). Overall prevalence of maternal CVD was 4% (age-adjusted 7%) and increased over 19 years of follow-up (age-adjusted prevalence of maternal CVD: 1% in 2001 to 7% in 2019, p <0.001). The incidence of pregnancy-related CV complications was 15% (age-adjusted 17%) and also increased over the follow-up period (age-adjusted incidence 11% in 2001 to 14% in 2019, p <0.001). Finally, CV complications were more likely to occur in individuals with greater burden of maternal CV comorbidities and CVD (diabetes: 6% vs 3%, hypertension: 23% vs 5%, pre-existing CVD: 10% vs 3%, P<0.001 for all).

Conclusions and Relevance

Analysis of a large-scale EHR-based pregnancy cohort spanning two decades demonstrates rising prevalence of both maternal cardiometabolic comorbidities and CVD at the time of pregnancy, as well as increasing incidence of subsequent pregnancy-related CV complications. Pregnancy represents a critical opportunity for cardiometabolic health optimization.

KEY POINTS

Question: What are the contemporary real-world trends in the prevalence of maternal cardiovascular comorbidities and cardiovascular disease and incidence of cardiovascular complications in pregnancy?Findings: In an analysis of 57,683 pregnancies among 38,997 individuals from a large scale EHR-based pregnancy cohort, prevalence of maternal cardiometabolic comorbidities and cardiovascular disease and incidence of pregnancy-related cardiovascular complications increased over the course of nearly two decades.Meaning: The contemporary burden of pregnancy-related cardiovascular complications is rising at an alarming rate and highlights pregnancy as a critical opportunity for cardiovascular health optimization.

Nature and prevalence of long-term conditions in people with intellectual disability: retrospective longitudinal population-based study

OBJECTIVE

Explore the nature and prevalence of long-term conditions in individuals with intellectual disability.

DESIGN

Retrospective longitudinal population-based study.

SETTING

Primary and secondary care data across the population of Wales with the Secure Anonymised Information Linkage (SAIL) Databank.

PARTICIPANTS

14 323 individuals were identified during the study date period 1 January 2000 to 31 December 2021 using the following inclusion criteria: 18 or older, alive at the cohort start date, a resident of Wales, with a primary care registration at a SAIL providing general practice with available records and a recorded diagnosis of intellectual disability. Once individuals were identified, health records were observed from birth.

RESULTS

13 069 individuals had a recorded diagnosis of intellectual disability and at least one long-term condition, reflecting 91.25% of the population. Demographic data from the SAIL dataset reveal that the study population is predominantly White, with low levels of representation of non-White ethnic groups. In the cohort, a larger proportion of patients live in the most deprived areas of Wales (22.30%), with fewer individuals in less deprived categories. Mental illness was identified as the most prevalent of the identified long-term conditions, whereby 30.91% of the population had a recorded diagnosis of a mental illness which was chronic. For many common conditions, including epilepsy, thyroid disorders, upper gastrointestinal disorders, chronic kidney disease and diabetes, there was an overall trend of higher prevalence rates in the intellectual disability cohort when compared with the general population. The prevalence of hypertension was lower in individuals with intellectual disability. Chronic constipation, chronic diarrhoea and insomnia were examples of long-term conditions added as relevant to individuals with intellectual disability. Notable differences in the distribution of long-term conditions were observed when comparing across sex and age groups. The number of long-term conditions increases with age. Conditions which may usually be expected to emerge later in life are present in younger age groups, such as diabetes, hypertension and chronic arthritis. When hospital episodes were analysed, epilepsy, diabetes, chronic airway disease and mental illness were commonly treated conditions during hospital admission across both sexes. Conditions which were less prevalent in the intellectual disability cohort, but which were treated during ≥6% of total hospital admissions include cancer, cardiac arrhythmias and cerebral palsy.

CONCLUSIONS

This study establishes a range of 40 relevant long-term conditions for people with intellectual disability through an iterative process, which included a review of the available literature and a series of discussions with a Professional Advisory Panel and Patient and Public Involvement groups of this research project. The findings of the study reinforce the high prevalence and early emergence of long-term conditions in the intellectual disability cohort. It also demonstrates the difference in the range of conditions when compared with the general population. There were differences in long-term conditions when separated by sex and age. Long-term conditions which commonly require treatment in hospitals were also revealed. Further work is required to translate the findings of this study into actionable insights. Clusters of multiple long-term conditions, trajectories, outcomes and risk factors should be explored to optimise the understanding and longitudinal care of individuals with intellectual disabilities and long-term conditions.

Large simple randomized controlled trials-from drugs to medical devices: lessons from recent experience

Randomized controlled trials (RCTs) are the cornerstone of modern evidence-based medicine. They are considered essential to establish definitive evidence of efficacy and safety for new drugs, and whenever possible they should also be the preferred method for investigating new high-risk medical devices. Well-designed studies robustly inform clinical practice guidelines and decision-making, but administrative obstacles have made it increasingly difficult to conduct informative RCTs. The obstacles are compounded for RCTs of high-risk medical devices by extra costs related to the interventional procedure that is needed to implant the device, challenges with willingness to randomize patients throughout a trial, and difficulties in ensuring proper blinding even with sham procedures. One strategy that may help is to promote the wider use of simpler and more streamlined RCTs using data that are collected routinely during healthcare delivery. Recent large simple RCTs have successfully compared the performance of drugs and of high-risk medical devices, against alternative treatments; they enrolled many patients in a short time, limited costs, and improved efficiency, while also achieving major impact. From a task conducted within the CORE-MD project, we report from our combined experience of designing and conducting large pharmaceutical trials during the COVID-19 pandemic, and of planning and coordinating large registry-based RCTs of cardiovascular devices. We summarize the essential principles and utility of large simple RCTs, likely applicable to all interventions but especially in order to promote their wider adoption to evaluate new medical devices.

Methodological challenges using routine clinical care data for real-world evidence: a rapid review utilizing a systematic literature search and focus group discussion

BACKGROUND

The integration of real-world evidence (RWE) from real-world data (RWD) in clinical research is crucial for bridging the gap between clinical trial results and real-world outcomes. Analyzing routinely collected data to generate clinical evidence faces methodological concerns like confounding and bias, similar to prospectively documented observational studies. This study focuses on additional limitations frequently reported in the literature, providing an overview of the challenges and biases inherent to analyzing routine clinical care data, including health claims data (hereafter: routine data).

METHODS

We conducted a literature search on routine data studies in four high-impact journals based on the Journal Citation Reports (JCR) category "Medicine, General & Internal" as of 2022 and three oncology journals, covering articles published from January 2018 to October 2023. Articles were screened and categorized into three scenarios based on their potential to provide meaningful RWE: (1) Burden of Disease, (2) Safety and Risk Group Analysis, and (3) Treatment Comparison. Limitations of this type of data cited in the discussion sections were extracted and classified according to different bias types: main bias categories in non-randomized studies (information bias, reporting bias, selection bias, confounding) and additional routine data-specific challenges (i.e., operationalization, coding, follow-up, missing data, validation, and data quality). These classifications were then ranked by relevance in a focus group meeting of methodological experts. The search was pre-specified and registered in PROSPERO (CRD42023477616).

RESULTS

In October 2023, 227 articles were identified, 69 were assessed for eligibility, and 39 were included in the review: 11 on the burden of disease, 17 on safety and risk group analysis, and 11 on treatment comparison. Besides typical biases in observational studies, we identified additional challenges specific to RWE frequently mentioned in the discussion sections. The focus group had varied opinions on the limitations of Safety and Risk Group Analysis and Treatment Comparison but agreed on the essential limitations for the Burden of Disease category.

CONCLUSION

This review provides a comprehensive overview of potential limitations and biases in analyzing routine data reported in recent high-impact journals. We highlighted key challenges that have high potential to impact analysis results, emphasizing the need for thorough consideration and discussion for meaningful inferences.

Exploring Mortality and Prognostic Factors of Heart Failure with In-Hospital and Emergency Patients by Electronic Medical Records: A Machine Learning Approach

Purpose

As HF progresses into advanced HF, patients experience a poor quality of life, distressing symptoms, intensive care use, social distress, and eventual hospital death. We aimed to investigate the relationship between morality and potential prognostic factors among in-patient and emergency patients with HF.

Patients and Methods

A case series study: Data are collected from in-hospital and emergency care patients from 2014 to 2021, including their international classification of disease at admission, and laboratory data such as blood count, liver and renal functions, lipid profile, and other biochemistry from the hospital's electrical medical records. After a series of data pre-processing in the electronic medical record system, several machine learning models were used to evaluate predictions of HF mortality. The outcomes of those potential risk factors were visualized by different statistical analyses.

Results

In total, 3871 hF patients were enrolled. Logistic regression showed that intensive care unit (ICU) history within 1 week (OR: 9.765, 95% CI: 6.65, 14.34; p-value < 0.001) and prothrombin time (OR: 1.193, 95% CI: 1.098, 1.296; <0.001) were associated with mortality. Similar results were obtained when we analyzed the data using Cox regression instead of logistic regression. Random forest, support vector machine (SVM), Adaboost, and logistic regression had better overall performances with areas under the receiver operating characteristic curve (AUROCs) of >0.87. Naïve Bayes was the best in terms of both specificity and precision. With ensemble learning, age, ICU history within 1 week, and respiratory rate (BF) were the top three compelling risk factors affecting mortality due to HF. To improve the explainability of the AI models, Shapley Additive Explanations methods were also conducted.

Conclusion

Exploring HF mortality and its patterns related to clinical risk factors by machine learning models can help physicians make appropriate decisions when monitoring HF patients' health quality in the hospital.

Sharing data matters: exploring the attitudes of older consumers on an emerging healthy ageing data platform using electronic health records for research

BACKGROUND

In Australia, with the recent introduction of electronic health records (EHRs) into hospitals, the use of hospital-based EHRs for research is a relatively new concept. The aim of this study was to explore the attitudes of older healthcare consumers on sharing their health data with an emerging EHR-based Research Data Platform within the National Centre for Healthy Ageing.

METHODS

This was a qualitative study. Two workshops were conducted in March 2022 with consumer representatives across Peninsula Health, Victoria, Australia. The workshops comprised three parts: (1) an ice-breaker (2) an introduction to EHR-based research through the presentation of 'use case' scenarios and (3) focus group discussions. Qualitative data were analysed using reflexive thematic analysis.

RESULTS

Consumer participants (n=16) were aged between 62 and 83 years and were of mixed gender. The overarching theme was related to trust in the use of EHR data for research; themes included: (1) benefits of sharing data, (2) uncertainty around data collection processes and (3) data sharing fears. The three themes within the overarching theme all reflect participants' levels of trust.

CONCLUSION

Our study identified fundamental issues related to trust in the use of EHR data for research, with both healthcare and broader societal factors contributing to consumer attitudes. Processes to support transparent and clear communication with consumers are essential to support the responsible use of EHR data for research.

Randomized in error in pragmatic clinical trials

BACKGROUND

Pragmatic trials that combine electronic health record data and patient-reported data may be subject to selection bias due to the differential post-randomization exclusion of participants who are randomized in error. Such situations are often caused by inevitable reasons, such as incomplete patient medical records at the pre-randomization stage. This can lead to participants in the intervention arm being identified as ineligible after randomization, while randomized-in-error participants in the usual care are often not discernable. The differential exclusion can present analytic challenges and threaten result validity.

METHODS

Under the potential outcomes framework, we developed a Bayesian model that jointly identifies the randomized-in-error status and estimates the average treatment effect among participants not randomized in error. We designed simulation studies with hypothesized proportions of 5 %-15 % randomization in error to evaluate the performance of our model across scenarios where the outcomes of participants randomized in error were either measured or unmeasured. Comparisons were made to intention-to-treat and covariate-adjusted estimators.

RESULTS

Simulation results show satisfactory performance of our proposed models, where the estimated average treatment effects among participants not randomized in error have low bias (<1 %) and close to 95 % coverage. Estimates from the alternative approaches can exhibit notable biases and low coverage.

CONCLUSIONS

Differential exclusion in pragmatic clinical trials after randomization can lead to selection bias. Under certain assumptions, Bayesian methods provide a feasible solution to jointly identify randomized-in-error status and estimate the average treatment effect among participants not randomized in error, ensuring more reliable and valid inferences about intervention effects.

LungDiag: Empowering artificial intelligence for respiratory diseases diagnosis based on electronic health records, a multicenter study

Respiratory diseases pose a significant global health burden, with challenges in early and accurate diagnosis due to overlapping clinical symptoms, which often leads to misdiagnosis or delayed treatment. To address this issue, we developed LungDiag, an artificial intelligence (AI)-based diagnostic system that utilizes natural language processing (NLP) to extract key clinical features from electronic health records (EHRs) for the accurate classification of respiratory diseases. This study employed a large cohort of 31,267 EHRs from multiple centers for model training and internal testing. Additionally, prospective real-world validation was conducted using 1142 EHRs from three external centers. LungDiag demonstrated superior diagnostic performance, achieving an F1 score of 0.711 for top 1 diagnosis and 0.927 for top 3 diagnoses. In real-world testing, LungDiag outperformed both human experts and ChatGPT 4.0, achieving an F1 score of 0.651 for top 1 diagnosis. The study emphasizes the potential of LungDiag as an effective tool to support physicians in diagnosing respiratory diseases more accurately and efficiently. Despite the promising results, further large-scale multicenter validation with larger sample sizes is still needed to confirm its clinical utility and generalizability.

Improving electronic health record documentation and use to promote evidence-based pediatric care

OBJECTIVE

Electronic health records (EHRs) often lack the necessary functionalities to support the full implementation of national clinical guidelines for pediatric care outlined in the American Academy of Pediatrics Bright Futures Guidelines. Using HealthySteps (HS), an evidence-based pediatric primary care program, as an exemplar, this study aimed to enhance pediatric EHRs, identify facilitators and barriers to EHR enhancements, and improve data quality for delivering clinical care as part of HS implementation and evidence building.

METHODS

Three HS sites-each differing in location, setting, number of children served, and mix of child insurance coverage-participated in the study. Sites received technical assistance to support data collection and EHR updates. A comprehensive evaluation, including a process evaluation and outcomes monitoring, was conducted to gauge progress toward implementing study data requirements over time. Data sources included administrative records, surveys, and interviews.

RESULTS

All sites enhanced their EHRs yet relied on supplemental data systems to track care coordination. Sites improved documentation of required data, demonstrating reductions in missing data and increases in extractable data between baseline and follow-up assessments. For example, the percentage of missing social-emotional screening results ranged from 0% to 8.0% at study conclusion. Facilitators and barriers to EHR enhancements included organizational supports, leadership, and capacity building.

CONCLUSIONS

With significant investment of time and resources, practices modified their EHRs to better capture services aligned with HS and Bright Futures. However, more scalable digital solutions are necessary to support EHR updates to help drive improvements in clinical care and outcomes for children and families.

Efficacy of antiobesity medications for weight reduction in older adults: a systematic review

OBJECTIVE

The objective of this study was to examine weight reduction and adverse events associated with use of antiobesity medications (AOMs) in older adults ages ≥65 years.

METHODS

Seven databases were searched for studies evaluating weight reduction of Food and Drug Administration (FDA)-approved AOMs. Studies had to include adults ages ≥65 years with obesity (BMI ≥ 30 kg/m2 or ≥27 kg/m2 with one weight-related condition), with independent analysis of weight reduction for adults ages ≥65 years. Two coauthors extracted and evaluated studies for risk of bias using standardized forms.

RESULTS

Six experimental studies (five secondary analyses of randomized clinical trial data and one single-arm trial) and two observational studies met inclusion criteria. Seven medications were studied. Sample size of older adults ranged from 13 to 6728. Experimental studies predominantly included patients with concurrent prediabetes or cardiovascular disease. All studies found statistically significant weight reduction between intervention and placebo groups or compared with baseline weight. Few studies reported on adverse events.

CONCLUSIONS

Limited evidence suggests weight reduction of AOMs in older adults, with the best current evidence for the use of semaglutide in older adults with obesity and cardiovascular disease. Larger, more inclusive studies of older adults are needed to guide clinical care and determine the tolerability of AOMs for older adults.

Medical Marijuana Documentation Practices in Patient Electronic Health Records: Retrospective Observational Study Using Smart Data Elements and a Review of Medical Records

Background

Medical marijuana (MMJ) is available in Pennsylvania, and participation in the state-regulated program requires patient registration and receiving certification by an approved physician. Currently, no integration of MMJ certification data with health records exists in Pennsylvania that would allow clinicians to rapidly identify patients using MMJ, as exists with other scheduled drugs. This absence of a formal data sharing structure necessitates tools aiding in consistent documentation practices to enable comprehensive patient care. Customized smart data elements (SDEs) were made available to clinicians at an integrated health system, Geisinger, following MMJ legalization in Pennsylvania.

Objective

The purpose of this project was to examine and contextualize the use of MMJ SDEs in the Geisinger population. We accomplished this goal by developing a systematic protocol for review of medical records and creating a tool that resulted in consistent human data extraction.

Methods

We developed a protocol for reviewing medical records for extracting MMJ-related information. The protocol was developed between August and December of 2022 and focused on a patient group that received one of several MMJ SDEs between January 25, 2019, and May 26, 2022. Characteristics were first identified on a pilot sample (n=5), which were then iteratively reviewed to optimize for consistency. Following the pilot, 2 reviewers were assigned 200 randomly selected patients' medical records, with a third reviewer examining a subsample (n=30) to determine reliability. We then summarized the clinician- and patient-level features from 156 medical records with a table-format SDE that best captured MMJ information.

Results

We found the review protocol for medical records was feasible for those with minimal medical background to complete, with high interrater reliability (κ=0.966; P<.001; odds ratio 0.97, 95% CI 0.954-0.978). MMJ certification was largely documented by nurses and medical assistants (n=138, 88.5%) and typically within primary care settings (n=107, 68.6%). The SDE has 6 preset field prompts with heterogeneous documentation completion rates, including certifying conditions (n=146, 93.6%), product (n=145, 92.9%), authorized dispensary (n=137, 87.8%), active ingredient (n=130, 83.3%), certifying provider (n=96, 61.5%), and dosage (n=48, 30.8%). We found preset fields were overall well-recorded (mean 76.6%, SD 23.7% across all fields). Primary diagnostic codes recorded at documentation encounters varied, with the most frequent being routine examinations and testing (n=34, 21.8%), musculoskeletal or nervous conditions, and signs and symptoms not classified elsewhere (n=21, 13.5%).

Conclusions

This method of reviewing medical records yields high-quality data extraction that can serve as a model for other health record inquiries. Our evaluation showed relatively high completeness of SDE fields, primarily by clinical staff responsible for rooming patients, with an overview of conditions under which MMJ is documented. Improving the adoption and fidelity of SDE data collection may present a valuable data source for future research on patient MMJ use, treatment efficacy, and outcomes.

Protocol for an embedded randomised controlled trial of Early versus Late Stopping of Antibiotics in children with Febrile Neutropenia (ELSA-FN)

In children with cancer, febrile neutropenia (FN) is one of the most common complications of treatment, a leading cause of unplanned and prolonged hospital admission and is the key driver of antibiotic exposure. Co-designed with key stakeholders, 'Early versus Late Stopping of Antibiotics in high-risk FN' (ELSA-FN) is a randomised controlled, non-inferiority trial that compares stopping antibiotics in clinically stable patients after 48 hours with the current standard of care, continuing antibiotics until absolute neutrophil recovery. As an Australian first, we will exploit the potential of electronic medical record (EMR) systems, embedding all key aspects of the trial including screening, consent, randomisation and data collection into standard clinical and EMR workflows. We aim to randomise 320 children with high-risk FN and prospectively collect data on safety, acceptability to clinicians and families, as well as several secondary outcomes related to antibiotic exposure. The findings will contribute to optimal antibiotic use in children with FN internationally and inform design and implementation of future EMR-embedded trials.

"The Truth Is, We Must Miss Some": A Qualitative Study of the Patient Eligibility Screening Process, and Automation Perspectives, for Cancer Clinical Trials

BACKGROUND

Recruitment of cancer patients into clinical trials (CTs) is a challenge. We aimed to explore how patient eligibility assessment is conducted in practice, what factors support or hinder this process, and to assess the potential usefulness of Clinical Trial Recruitment Support Systems (CTRSS) for patient-to-trial matching.

METHODS

We conducted semi-structured interviews in France with healthcare professionals involved in cancer CTs and experts on trial recruitment. We focused on the stages in-between trial feasibility, and patient information and consent. Interviews were recorded, and the transcripts were analyzed thematically. We used the Systems Engineering Initiative for Patient Safety (SEIPS) 2.0 framework to organize our results.

RESULTS

We interviewed 25 participants. We identified common steps for cancer patient eligibility assessment: prescreening under medical supervision, followed by the validation of patient-trial matching based on manual chart review. This process built on rich interactions between clinicians, other professionals (clinical research assistants, data scientists, medical coding experts), and patients. Technological factors, mainly related to data infrastructure (both for patient data and trial data), and organizational factors (research culture, incentives, formal and informal research networks) mediated the performance of the recruitment process. Participants had mixed feelings towards CTRSSs; they welcomed automated pre-screening but insisted on manual verification. Given the necessary collaborative nature of multisite trials, coordinated efforts to support a common data infrastructure could be helpful.

CONCLUSIONS

Material, organizational, and human factors affect cancer patient eligibility assessment for CTs. Patient-to-trial matching tools bear potential, but good understanding of the ecosystem, including stakeholders' motivations, is a prerequisite.

Optimizing Home Visit Records as a Way of Improving Quality of Care: A Quality Improvement Study

Introduction Home visits are a key component of primary care in Portugal, designed for patients unable to visit medical facilities. However, logistical constraints often lead to incomplete real-time clinical records, impacting care quality and safety. This study aimed to improve the quality of home visit records through structural interventions and a continuous quality improvement approach. Methods This study was conducted in a Portuguese family health unit between February and December 2023. This retrospective study involved all home visits performed by physicians from October 2022 to October 2023. Using the Plan-Do-Study-Act (PDSA) methodology, records were assessed based on four parameters: accuracy of the "Assessment" section of the Subjective, Objective, Assessment, and Plan (SOAP) note (aligned with the International Classification of Primary Care, 2nd edition); Barthel scale documentation; updated list of problems; and updated list of chronic medication. Data were collected, analyzed descriptively, and presented at three time points: baseline evaluation (February 2023), intermediate evaluation (July 2023), and post-intervention evaluation (December 2023). Two interventions were made, including educational sessions and the introduction of physical support tools for record-keeping. The established quality-defining goal was to achieve compliance with all four parameters in at least 80% of records. Results At baseline, none of the 97 evaluated records met all criteria. After two interventions, compliance significantly improved. By the final evaluation, 74.7% of 95 records met all criteria, while no records failed entirely. Discussion Despite not fully achieving the predefined goal, interventions significantly enhanced record quality, ranging from 0% to 74.7% at the end of the study. These findings demonstrate the value of structural interventions and collaborative team efforts in improving home visit records. Despite significant progress in improving home visit records, there is still room for improvement. It is essential for healthcare professionals to continue enhancing record-keeping practices to improve the effectiveness of domiciliary care and patient outcomes. Conclusion This study highlights the importance of accurate clinical records for safe and effective domiciliary care. Continued commitment to structured record-keeping practices and further research is essential to sustain improvements and optimize patient outcomes.

Real-World Evidence on Adverse Events and Healthcare Resource Utilization in Patients with Chronic Lymphocytic Leukaemia in Spain Using Natural Language Processing: The SRealCLL Study

Objectives: The SRealCLL study described the occurrence of adverse events (AEs) and healthcare resource utilization in patients with chronic lymphocytic leukaemia (CLL) using artificial intelligence in a real-world scenario in Spain. Methods: We collected real-world data on patients with CLL from seven Spanish hospitals between January 2016 and December 2018, focusing on their AE and healthcare service utilization. Data extraction from electronic health records of 385,904 patients was performed using the EHRead® technology, which is based on natural language processing and machine learning. Results: Among the 534 CLL patients finally included, 270 (50.6%) were categorized as watch and wait (W&W), 230 (43.1%) as first-line treatment (1L), and 58 (10.9%) as relapse/refractory with second-line treatment (2L). The median study follow-up periods were 14.4, 8.4, and 6 months for W&W, 1L, and 2L, respectively. The most common antineoplastic treatments were ibrutinib (64.8%) and bendamustine + rituximab (12.6%) in 1L, and ibrutinib (62.1%) and venetoclax (15.5%) in 2L. Among the most frequent AEs, anaemia and thrombocytopenia presented higher rates in the treated groups (1L and 2L) compared with W&W (2.01 and 2.32 vs. 0.93; p ≤ 0.05 and 1.29 and 1.62 vs. 0.42; p ≤ 0.05). Moreover, several AEs, such as major bleeding, digestive symptoms, general symptoms, or Richter syndrome, were more frequent in 1L than W&W (all p ≤ 0.05). No differences were shown between groups in the rates of outpatient visits. However, rates of outpatient visits due to AE were higher in 1L than in W&W (1.07 vs. 0.65, p ≤ 0.05). The rates of patients being hospitalized were higher in the treated groups compared to W&W (1.68 and 1.9 vs. 0.88; p ≤ 0.05), and those due to AE were higher in 1L than W&W (1.23 vs. 0.60; p ≤ 0.05). Conclusions: Patients with CLL in 1L or 2L treatments often require healthcare resources due to AEs, particularly cytopenias. The methodology used in this study likely enabled us to identify higher rates of AEs that may be underreported using other real-world approaches. Addressing AEs with effective agents that maximize patient safety and optimize healthcare resource use is crucial in this typically older and comorbid population.

Anomaly-based threat detection in smart health using machine learning

BACKGROUND

Anomaly detection is crucial in healthcare data due to challenges associated with the integration of smart technologies and healthcare. Anomaly in electronic health record can be associated with an insider trying to access and manipulate the data. This article focuses around the anomalies under different contexts.

METHODOLOGY

This research has proposed methodology to secure Electronic Health Records (EHRs) within a complex environment. We have employed a systematic approach encompassing data preprocessing, labeling, modeling, and evaluation. Anomalies are not labelled thus a mechanism is required that predicts them with greater accuracy and less false positive results. This research utilized unsupervised machine learning algorithms that includes Isolation Forest and Local Outlier Factor clustering algorithms. By calculating anomaly scores and validating clustering through metrics like the Silhouette Score and Dunn Score, we enhanced the capacity to secure sensitive healthcare data evolving digital threats. Three variations of Isolation Forest (IForest)models (SVM, Decision Tree, and Random Forest) and three variations of Local Outlier Factor (LOF) models (SVM, Decision Tree, and Random Forest) are evaluated based on accuracy, sensitivity, specificity, and F1 Score.

RESULTS

Isolation Forest SVM achieves the highest accuracy of 99.21%, high sensitivity (99.75%) and specificity (99.32%), and a commendable F1 Score of 98.72%. The Isolation Forest Decision Tree also performs well with an accuracy of 98.92% and an F1 Score of 99.35%. However, the Isolation Forest Random Forest exhibits lower specificity (72.84%) than the other models.

CONCLUSION

The experimental results reveal that Isolation Forest SVM emerges as the top performer showcasing the effectiveness of these models in anomaly detection tasks. The proposed methodology utilizing isolation forest and SVM produced better results by detecting anomalies with less false positives in this specific EHR of a hospital in North England. Furthermore the proposal is also able to identify new contextual anomalies that were not identified in the baseline methodology.

Multisource representation learning for pediatric knowledge extraction from electronic health records

Electronic Health Record (EHR) systems are particularly valuable in pediatrics due to high barriers in clinical studies, but pediatric EHR data often suffer from low content density. Existing EHR code embeddings tailored for the general patient population fail to address the unique needs of pediatric patients. To bridge this gap, we introduce a transfer learning approach, MUltisource Graph Synthesis (MUGS), aimed at accurate knowledge extraction and relation detection in pediatric contexts. MUGS integrates graphical data from both pediatric and general EHR systems, along with hierarchical medical ontologies, to create embeddings that adaptively capture both the homogeneity and heterogeneity between hospital systems. These embeddings enable refined EHR feature engineering and nuanced patient profiling, proving particularly effective in identifying pediatric patients similar to specific profiles, with a focus on pulmonary hypertension (PH). MUGS embeddings, resistant to negative transfer, outperform other benchmark methods in multiple applications, advancing evidence-based pediatric research.

Artificial Intelligence in Cardiovascular Clinical Trials

Randomized clinical trials are the gold standard for establishing the efficacy and safety of cardiovascular therapies. However, current pivotal trials are expensive, lengthy, and insufficiently diverse. Emerging artificial intelligence (AI) technologies can potentially automate and streamline clinical trial operations. This review describes opportunities to integrate AI throughout a trial's life cycle, including designing the trial, identifying eligible patients, obtaining informed consent, ascertaining physiological and clinical event outcomes, interpreting imaging, and analyzing or disseminating the results. Nevertheless, AI poses risks, including generating inaccurate results, amplifying biases against underrepresented groups, and violating patient privacy. Medical journals and regulators are developing new frameworks to evaluate AI research tools and the data they generate. Given the high-stakes role of randomized trials in medical decision making, AI must be integrated carefully and transparently to protect the validity of trial results.

Development and validation of a novel AI framework using NLP with LLM integration for relevant clinical data extraction through automated chart review

The accurate extraction of surgical data from electronic health records (EHRs), particularly operative notes through manual chart review (MCR), is complex, crucial, and time-intensive, limited by human error due to fatigue and the level of training. This study aimed to develop and validate a novel Natural Language Processing (NLP) algorithm integrated with a Large Language Model (LLM; GPT4-Turbo) to automate the extraction of spinal surgery data from EHRs. The algorithm employed a two-stage approach. Initially, a rule-based NLP framework reviewed and classified candidate segments from the text, preserving their reference segments. These segments were then verified in the second stage through the LLM. The primary outcomes of this study were the accurate extraction of surgical data, including the type of surgery, levels operated, number of disks removed, and presence of intraoperative incidental durotomies. Secondary objectives explored time efficiency, tokenization lengths, and costs. The performance of the algorithm was assessed across two validation databases, analyzing metrics such as accuracy, sensitivity, discrimination, F1-score, and precision, with 95% confidence intervals calculated using percentile-based bootstrapping. The NLP + LLM algorithm markedly outperformed all performance metrics, demonstrating significant improvements in time and cost efficiency. These results suggest the potential for widespread adoption of this technology.

Automated Electronic Health Record Data Extraction and Curation Using ExtractEHR

PURPOSE

Although the potential transformative effect of electronic health record (EHR) data on clinical research in adult patient populations has been very extensively discussed, the effect on pediatric oncology research has been limited. Multiple factors contribute to this more limited effect, including the paucity of pediatric cancer cases in commercial EHR-derived cancer data sets and phenotypic case identification challenges in pediatric federated EHR data.

METHODS

The ExtractEHR software package was initially developed as a tool to improve clinical trial adverse event reporting but has expanded its use cases to include the development of multisite EHR data sets and the support of cancer cohorts. ExtractEHR enables customized, automated data extraction from the EHR that, when implemented across multiple hospitals, can create pediatric cancer EHR data sets to address a very wide range of research questions in pediatric oncology. After ExtractEHR data acquisition, EHR data can be cleaned and graded using CleanEHR and GradeEHR, companion software packages.

RESULTS

ExtractEHR has been installed at four leading pediatric institutions: Children's Healthcare of Atlanta, Children's Hospital of Philadelphia, Texas Children's Hospital, and Seattle Children's Hospital.

CONCLUSION

ExtractEHR has supported multiple use cases, including five clinical epidemiology studies, multicenter clinical trials, and cancer cohort assembly. Work is ongoing to develop Fast Health care Interoperability Resources ExtractEHR and implement other sustainability and scalability enhancements.

Machine Learning on 50,000 Manuscripts Shows Increased Clinical Research by Academic Cardiac Surgeons

INTRODUCTION

Academic cardiac surgeons are productive researchers and innovators. We sought to perform a comprehensive machine learning (ML)-based characterization of cardiac surgery research over the past 40 y to identify trends in research pursuits.

METHODS

US-based academic websites were queried for surgeon profiles. Publications since 1980 were obtained from Web of Science, and publication classifications (e.g., "human", "animal") were collected through the National Institutes of Health iCite tool. Publications were deemed "basic or translational" if >50% of their classification was under "animal" or "molecular or cell", and "clinical" if otherwise. ML-based clustering was performed on publication titles and Medical Subject Heading terms to identify research topics.

RESULTS

A total of 944 cardiac surgeons accounted for 48,031 unique publications. Average citations per year have decreased since 1980 (P < 0.001). The percentage of basic or translational publications by cardiac surgeons has decreased over time (P < 0.001), comprising of only 8% of publications in 2022. Adult cardiac surgeons, those who received an F32, K08, or R01, and those with a PhD were more likely to publish basic or translational research. Top areas of basic or translational research were myocardial reperfusion, aortic aneurysms or remodeling, and transplant immunology. Major areas of clinical research included aortic disease, aortic valve disease, and mechanical circulatory support. Collaboration analysis revealed that 55% of publications were single-center, and the yearly percentage of these publications has decreased over time (P < 0.001).

CONCLUSIONS

Cardiac surgeons are performing less basic or translational research relative to clinical research than ever before. The majority of publications over the past 40 y did not involve cross-center collaboration. Continued support for clinical research is needed, while also encouraging collaborative basic or translational science to foster innovation in patient care.

MAPRS: An intelligent approach for post-prescription review based on multi-label learning

Antimicrobial resistance (AMR) is a major threat to public health worldwide. It is a promising way to improve appropriate prescription by the review and stewardship of antimicrobials, and Post-Prescription Review (PPR) is currently the main tool used in hospitals. Existing methods of PPR typically focus on the dichotomy of antimicrobial prescription based on binary classification which, however, is usually a multi-label classification problem. Moreover, previous research did not explain the causes beneath the inappropriate antimicrobial used in the clinical setting, which could be practically important for problem location and decision improvement. In this paper, we collected antimicrobial prescriptions and related data from clean surgery in a hospital in northeastern China, and proposed a Multi-label Antimicrobial Post-Prescription Review System (MAPRS). MAPRS first uses NLP techniques to process unstructured data in prescriptions and explores the value of clinical record text for solving medical problems. Then, Classifier Chains are used to deal with multi-label problems and fused with machine learning algorithms to construct a classifier. At last, a SHAP explanation module is introduced to explain the inappropriate prescriptions. The experimental results show that MAPRS could achieve great performance in a challenging six-category multi-label task, with a subset accuracy of 90.7 % and an average AUROC of 94.3 %. Our results can help hospitals to perform intelligent prescription review and improve the antimicrobial stewardship.

Information Management for the Neurosurgical Care of Brain Tumors: A Scoping Review of Literature from Low- and Middle-Income Countries

BACKGROUND

Health care in developing countries often lacks adequate bookkeeping and national cancer registries, means of information that have proven to impact disease research and care. The true burden of brain tumors therefore remains unchecked and so does the extent of the problem. Therefore, this study aims to explore the challenges and potential strategies related to information management of brain tumors in low- and middle-income countries (LMICs).

METHODS

A comprehensive literature search conducted using databases such as PubMed, Scopus, Google Scholar, and Cumulated Index in Nursing and Allied Health Literature, without any language restrictions, from inception to October 20, 2022. Following screening and extraction of data, themes were generated using the information management domain of the National Surgical, Obstetric, and Anesthesia Plan framework.

RESULTS

The final analysis includes 23 studies that highlighted the challenges to managing information to the surgical care given to brain tumors in LMICs, including lack of proper hospital record system (43%), lack of national brain tumor registry (67%), lack of local management guidelines (10%), and low research output (33%). Some of the proposed strategies in the literature to address these barriers include improving data management systems (45%), developing a population-based brain tumor registry (64%), and formulating local treatment guidelines (9%) for the management of brain tumors.

CONCLUSIONS

In LMICs, improving patient outcomes and quality of life postneurosurgical intervention for brain tumors requires coordinated efforts to enhance information systems. The support of the government and public health professionals is vital in implementing strategies to achieve this goal.

Development and validation of an open-source pipeline for automatic population of case report forms from electronic health records: a pediatric multi-center prospective study

BACKGROUND

Clinical trials and registry studies are essential for advancing research and developing novel treatments. However, these studies rely on manual entry of thousands of variables for each patient. Repurposing real-world data can significantly simplify the data collection, reduce transcription errors, and make the data entry process more efficient, consistent, and cost-effective.

METHODS

We developed an open-source computational pipeline to collect laboratory and medication information from the electronic health record (EHR) data and populate case report forms. The pipeline was developed and validated with data from two independent pediatric hospitals in the US as part of the Long-terM OUtcomes after Multisystem Inflammatory Syndrome In Children (MUSIC) study. Our pipeline allowed the completion of two of the most time-consuming forms. We compared automatically extracted results with manually entered values in one hospital and applied the pipeline to a second hospital, where the output served as the primary data source for case report forms.

FINDINGS

We extracted and populated 51,845 laboratory and 4913 medication values for 159 patients in two hospitals participating in a prospective pediatric study. We evaluated pipeline performance against data for 104 patients manually entered by clinicians in one of the hospitals. The highest concordance was found during patient hospitalization, with 91.59% of the automatically extracted laboratory and medication values corresponding with the manually entered values. In addition to the successfully populated values, we identified an additional 13,396 laboratory and 567 medication values of interest for the study.

INTERPRETATION

The automatic data entry of laboratory and medication values during admission is feasible and has a high concordance with the manually entered data. By implementing this proof of concept, we demonstrate the quality of automatic data extraction and highlight the potential of secondary use of EHR data to advance medical science by improving data entry efficiency and expediting clinical research.

FUNDING

NIH Grant 1OT3HL147154-01, U24HL135691, UG1HL135685.

Communication, Shared Decision-making and Goals of Care in the ICU through Electronic Health Records: A Scoping Review

Introduction

The care of critically ill patients involves communication and shared decision-making with families and determination of goals of care. Analyzing these aspects through electronic health records (EHRs) can support research in ICUs, associating them with outcomes. This review aims to explore studies that examine these topics.

Methods

A scoping review was conducted through a systematic literature search of articles in PubMed, Web of Science, and Embase databases using MESH terms up to 2024, conducted in ICU settings, focusing on communication with families, shared decision-making, goals, and end-of-life care.

Results

A total of 10 articles were included, divided into themes: Records and family, and records in quality improvement projects. Variables based on records with common characteristics were identified. Outcome analysis was performed through questionnaires to family members, healthcare professionals or by analyzing care processes. The studies revealed associations between family members' perceptions and mental health symptoms and documented elements such as communication, therapeutic limitations, social and spiritual support. Studies evaluating quality communication improvement projects did not show significant impact on documented care, except for those that assessed improvements based on palliative care.

Conclusion

The analysis of documented care for critically ill patients can be conducted from various perspectives. Processes amenable to improvement, such as communication with family members, definition of goals of care, limitations, shared decision-making, evaluated through EHRs, are associated with mental health symptoms and perceptions of families of critically ill patients. Documentation-based studies can contribute to improvements in patient- and family-centered care in the ICU.

How to cite this article

de Aquino VW, da Silveira GF, Boniatti MM, Terres MS. Communication, Shared Decision-making and Goals of Care in the ICU through Electronic Health Records: A Scoping Review. Indian J Crit Care Med 2024;28(10):977-987.

Digital approaches in post-COVID healthcare: a systematic review of technological innovations in disease management

Post-COVID conditions (PCC) emerged during the pandemic, prompting a rise in the use of Digital Health Technologies (DHTs) to manage lockdowns and hospital overcrowding. Real-time tracking and information analyses were crucial to strengthening the global research response. This study aims to map the use of modern digital approaches in estimating the prevalence, predicting, diagnosing, treating, monitoring, and prognosis of PCC. This review was conducted by searching PubMed and Scopus databases for keywords and synonyms related to DHTs, Smart Healthcare Systems, and PCC based on the World Health Organization definition. Articles published from 1 January 2020 to 21 May 2024 were screened for eligibility based on predefined inclusion criteria, and the PRISMA framework was used to report the findings from the retained studies. Our search identified 377 studies, but we retained 23 studies that used DHTs, artificial intelligence (AI), and infodemiology to diagnose, estimate prevalence, predict, treat, and monitor PCC. Notably, a few interventions used infodemics to identify the clinical presentations of the disease, while most utilized Electronic Health Records and AI tools to estimate diagnosis and prevalence. However, we found that AI tools were scarcely used for monitoring symptoms, and studies involving SHS were non-existent in low- and middle-income countries (LMICs). These findings show several DHTs used in healthcare, but there is an urgent need for further research in SHS for complex health conditions, particularly in LMICs. Enhancing DHTs and integrating AI and infodemiology provide promising avenues for managing epidemics and related complications, such as PCC.

Mapping the Clinical Pathway for Patients Undergoing Vestibular Schwannoma Resection

BACKGROUND

The introduction of the electronic health record (EHR) has improved the collection and storage of patient information, enhancing clinical communication and academic research. However, EHRs are limited by data quality and the time-consuming task of manual data extraction. This study aimed to use process mapping to help identify critical data entry points within the clinical pathway for patients with vestibular schwannoma (VS) ideal for structured data entry and automated data collection to improve patient care and research.

METHODS

A 2-stage methodology was used at a neurosurgical unit. Process maps were developed using semi-structured interviews with stakeholders in the management of VS resection. Process maps were then retrospectively validated against EHRs for patients admitted between August 2019 and December 2021, establishing critical data entry points.

RESULTS

In the process map development, 20 stakeholders were interviewed. Process maps were validated against EHRs of 36 patients admitted for VS resection. Operative notes, surgical inpatient reviews (including ward rounds), and discharge summaries were available for all patients, representing critical data entry points. Areas for documentation improvement were in the preoperative clinics (30/36; 83.3%), preoperative skull base multidisciplinary team (32/36; 88.9%), postoperative follow-up clinics (32/36; 88.9%), and postoperative skull base multidisciplinary team meeting (29/36; 80.6%).

CONCLUSIONS

This is a first use to our knowledge of a 2-stage methodology for process mapping the clinical pathway for patients undergoing VS resection. We identified critical data entry points that can be targeted for structured data entry and for automated data collection tools, positively impacting patient care and research.

Methods for pragmatic randomized clinical trials of pain therapies: IMMPACT statement

ABSTRACT

Pragmatic, randomized, controlled trials hold the potential to directly inform clinical decision making and health policy regarding the treatment of people experiencing pain. Pragmatic trials are designed to replicate or are embedded within routine clinical care and are increasingly valued to bridge the gap between trial research and clinical practice, especially in multidimensional conditions, such as pain and in nonpharmacological intervention research. To maximize the potential of pragmatic trials in pain research, the careful consideration of each methodological decision is required. Trials aligned with routine practice pose several challenges, such as determining and enrolling appropriate study participants, deciding on the appropriate level of flexibility in treatment delivery, integrating information on concomitant treatments and adherence, and choosing comparator conditions and outcome measures. Ensuring data quality in real-world clinical settings is another challenging goal. Furthermore, current trials in the field would benefit from analysis methods that allow for a differentiated understanding of effects across patient subgroups and improved reporting of methods and context, which is required to assess the generalizability of findings. At the same time, a range of novel methodological approaches provide opportunities for enhanced efficiency and relevance of pragmatic trials to stakeholders and clinical decision making. In this study, best-practice considerations for these and other concerns in pragmatic trials of pain treatments are offered and a number of promising solutions discussed. The basis of these recommendations was an Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMMPACT) meeting organized by the Analgesic, Anesthetic, and Addiction Clinical Trial Translations, Innovations, Opportunities, and Networks.

Cloud Horizons: Strengthening Rural Healthcare Through Telemedicine's Digital Canopy

Introduction

Cloud-based telemedicine holds promise for improving healthcare accessibility and delivery, particularly in rural areas of developing countries like Tanzania. However, little is known about its determinants and benefits in such contexts. This study investigates the factors influencing the usage of telemedicine in Mvomero district, Morogoro region, Tanzania, focusing on both supply and demand sides.

Method

Using structured interviews and key informant interviews, the study examines various cloud-based telemedicine platforms, including remote monitoring, electronic health records, cloud-based storage, and machine learning algorithms. The study used descriptive statistics to analyze quantitative data, while thematic analysis was used to analyze qualitative data.

Results

Results reveal several factors influencing telemedicine usage. On the demand side, perceived benefits (53.96%), technology cost (62.79%), legal practices (62.79%), and resource availability and affordability (49.77%) are crucial. On the supply side, technological innovation (35%) and access to financial resources (43%) play pivotal roles. Environmental and institutional factors such as political willingness (38%) and regulatory support (34%) also impact telemedicine usage. Moreover, results reveal that cloud-based telemedicine platforms in rural healthcare facilities have several benefits including improved access (32.74% to 57.44%), cost efficiency (37.88% to 54.82%), timely consultations (56.83% to 65.21%), health monitoring, and prescription management (43.89% to 75.90%). Private facilities particularly emphasize health monitoring.

Conclusion

Adopting telemedicine technologies can revolutionize rural healthcare by providing customized and easily accessible services. Policymakers can use these findings to develop targeted strategies, including subsidized infrastructure, innovative financing models, and clear regulatory frameworks. Clear guidelines on data transfer and privacy are essential to ensure legal compliance and equitable access to telemedicine benefits. Simplifying registration requirements and implementing explicit consent mechanisms are recommended to address data privacy concerns. These measures aim to promote operational efficiency, data safety, and enhanced health outcomes in resource-limited settings.

Comprehensive review of Korean Medicine registries 2015-2023

Background

Despite the increasing popularity of Korean Medicine (KM), its scientific evidence faces scrutiny. Instead of randomized controlled trials, registries are favored to capture the real world of KM practice due to the difficulties associated with proper control and the holistic nature of the KM approach. This review aimed to examine the KM registries in detail, identify the scope and focus of studies within this field, and assess the research trends.

Methods

We conducted a comprehensive analysis of KM registries listed in trial registration platforms, covering records from their inception until the end of 2023. The selection criteria aimed to include studies focusing on various interventions related to KM, with data extraction focusing on study characteristics and outcomes measured. The analysis utilized descriptive statistics to summarize the findings.

Results

We identified a steady increase in registry studies (2015, one; 2023, seven). Musculoskeletal disorders were most studied (28%), aligning with patients' demand. The involvement of 112 primary clinics and Quality of Life (QOL) as the predominant outcome in 14 (66.7%) registries demonstrates the positive impact on patient well-being and the critical role that primary clinics play in KM practice.

Conclusion

Our findings indicate a heightened interest and commitment to evidence-based KM practices. Future Registries should be implemented on a large scale, incorporating long-term follow-up encompassing primary clinics. This approach would enable a comprehensive evaluation of the effectiveness and safety of KM interventions, as well as offer valuable insights into the influence of KM on chronic conditions and QOL.