Electronic Healthcare Record Research Articles

S3EF-HBCAs: Secure and Sustainable Software Engineering Framework for Healthcare Blockchain Applications.

Blockchain applications in healthcare have grown rapidly. They include record-keeping, clinical trials, medical supply chains, patient monitoring, etc., where blockchain characteristics are needed to improve safety, privacy, and security. Blockchain technology is one of the most significant disruptive technologies today. However, Porru et al.1 reported that it lacks processes, tools, and techniques. Therefore, this paper provides a systematic framework for a secure and sustainable software engineering framework for healthcare blockchain applications (S3EF-HBCA). S3EF-HBCA is a significant contribution that includes requirements engineering for healthcare, business process modeling for healthcare, domain modeling for healthcare, a reference architecture for healthcare, and validation by a case study on electronic healthcare record management system (EHR), and simulation with business process modeling notation (BPMN) tools. The simulation shows it has taken 10.45 min to process 100 instances of real-time data and service requests. The overall result shows encouragement regarding process, tools, standards, and testing.

Axiomatizing SNOMED CT Disorders: Should There Be Room for Interpretation?

SNOMED CT is a large concept-based terminology designed according to epistemic, semantic and pragmatic principles relevant to clinicians. Its goal is structured clinical reporting in electronic healthcare records (EHRs). The Basic Formal Ontology (BFO) is an ontology designed on the basis of types claimed to exist in reality based on a domain-independent ontological theory. Its goal is faithful representation of reality within that theory. The Ontology for General Medical Science (OGMS) extends the BFO by providing definitions for types relevant within the clinical domain. Combining SNOMED CT with the ontological rigor of BFO and OGMS might improve clinical reporting by, f.i., preventing data entry mistakes and inconsistencies, and make EHRs more comparable. To that end, we are developing a logical framework capable of exploiting what SNOMED CT offers terminologically and realism-based ontologies such as the BFO and the OGMS ontologically by means of bridging axioms compatible with the BFO, and expressed in the same CLIF-dialect as used in its axiomatization in first order logic. In this paper, we report on our attempts to detect in the combinations of binary relations that are used in the definition of SNOMED CT's definitions of disorder concepts patterns which might at least partially automate the construction of such axioms. Our findings suggest that this partial automation is indeed possible, but to a smaller extent than we had hoped for. We compare our approach with a recent proposal that seeks to bring SNOMED CT and BFO closer together by reinterpreting SNOMED CT disorders as clinical occurrents. The proposal has its merit in providing a realist underpinning for that part of SNOMED CT's concept model in terms of the BFO, but is not discriminatory enough for an automatic translation into OGMS. Key problem is the lack of face validity of SNOMED CT disorder terms as compared to the formal definitions they are given and this in absence of textual definitions.

Phenotypes and rates of cancer-relevant symptoms and tests in the year before cancer diagnosis in UK Biobank and CPRD Gold.

Early diagnosis of cancer relies on accurate assessment of cancer risk in patients presenting with symptoms, when screening is not appropriate. But recorded symptoms in cancer patients pre-diagnosis may vary between different sources of electronic health records (EHRs), either genuinely or due to differential completeness of symptom recording. To assess possible differences, we analysed primary care EHRs in the year pre-diagnosis of cancer in UK Biobank and Clinical Practice Research Datalink (CPRD) populations linked to cancer registry data. We developed harmonised phenotypes in Read v2 and CTV3 coding systems for 21 symptoms and eight blood tests relevant to cancer diagnosis. Among 22,601 CPRD and 11,594 UK Biobank cancer patients, 54% and 36%, respectively, had at least one consultation for possible cancer symptoms recorded in the year before their diagnosis. Adjusted comparisons between datasets were made using multivariable Poisson models, comparing rates of symptoms/tests in CPRD against expected rates if cancer site-age-sex-deprivation associations were the same as in UK Biobank. UK Biobank cancer patients compared with those in CPRD had lower rates of consultation for possible cancer symptoms [RR: 0.61 (0.59-0.63)], and lower rates for any primary care consultation [RR: 0.86 (95%CI 0.85-0.87)]. Differences were larger for 'non-alarm' symptoms [RR: 0.54 (0.52-0.56)], and smaller for 'alarm' symptoms [RR: 0.80 (0.76-0.84)] and blood tests [RR: 0.93 (0.90-0.95)]. In the CPRD cohort, approximately representative of the UK population, half of cancer patients had recorded symptoms in the year before diagnosis. The frequency of non-specific presenting symptoms recorded in the year pre-diagnosis of cancer was substantially lower among UK Biobank participants. The degree to which results based on highly selected biobank cohorts are generalisable needs to be examined in disease-specific contexts.

Replicability and reproducibility of predictive models for diagnosis of depression among young adults using Electronic Health Records

BackgroundRecent advances in machine learning combined with the growing availability of digitized health records offer new opportunities for improving early diagnosis of depression. An emerging body of research shows that Electronic Health Records can be used to accurately predict cases of depression on the basis of individual’s primary care records. The successes of these studies are undeniable, but there is a growing concern that their results may not be replicable, which could cast doubt on their clinical usefulness.MethodsTo address this issue in the present paper, we set out to reproduce and replicate the work by Nichols et al. (2018), who trained predictive models of depression among young adults using Electronic Healthcare Records. Our contribution consists of three parts. First, we attempt to replicate the methodology used by the original authors, acquiring a more up-to-date set of primary health care records to the same specification and reproducing their data processing and analysis. Second, we test models presented in the original paper on our own data, thus providing out-of-sample prediction of the predictive models. Third, we extend past work by considering several novel machine-learning approaches in an attempt to improve the predictive accuracy achieved in the original work.ResultsIn summary, our results demonstrate that the work of Nichols et al. is largely reproducible and replicable. This was the case both for the replication of the original model and the out-of-sample replication applying NRCBM coefficients to our new EHRs data. Although alternative predictive models did not improve model performance over standard logistic regression, our results indicate that stepwise variable selection is not stable even in the case of large data sets.ConclusionWe discuss the challenges associated with the research on mental health and Electronic Health Records, including the need to produce interpretable and robust models. We demonstrated some potential issues associated with the reliance on EHRs, including changes in the regulations and guidelines (such as the QOF guidelines in the UK) and reliance on visits to GP as a predictor of specific disorders.

Clinical Features and Prognostic Factors in Multiple Myeloma Patients from Colombia. Real-World Data from Renehoc Registry

Background: The prognosis for patients with multiple myeloma (MM) has improved over time, but there is considerable heterogeneity in disease outcomes. Currently used risk models have been validated using data from patients in clinical trials, but their applicability to real-world populations may be limited. To provide clinicians with better tools for decision-making, it is essential to identify factors associated with overall survival (OS) in a local setting. This study aims to identify clinical factors associated with OS in a real-world cohort of Colombian patients with newly diagnosed multiple myeloma (NDMM) using data from the registry on hematologic diseases (RENEHOC). Methods: RENEHOC is an electronic healthcare record (EHR) system comprising data from 25 active centers in 6 Colombian cities as of July 2023. The study population was characterized using descriptive methodology. Two cohorts of patients were defined: Cohort A included patients who died within 5 years after diagnosis, and Cohort B included patients who were either alive or censored 5 years after diagnosis. Variables with a high proportion of missing information were evaluated in terms of magnitude and proportion. Univariate logistic regression was used to identify variables associated with OS, with those having a P value less than 0.2 considered for multivariate analysis. The results were presented as odds ratios (OR) with 95% confidence intervals (CI), and P values were calculated using the LR-test. Results: The analysis included 1579 patients with NDMM. The median age was 67 years (IQR 60-75), and 51.1% (n=808) of the patients were male. Most patients were insured through the contributory health system plan. At diagnosis, 36.9% (n=582) of patients had a bone fracture (n=425) and 26.9% (n=425) had renal insufficiency, with 9.8% requiring dialysis. Cytogenetic analysis was performed in 22.9% (n=361) of patients, and 24.1% (n=87) of them had high-risk features. Half of the patients were classified as ISS III at diagnosis. The median OS for the entire cohort was 98.4 months. According to the international staging system (ISS), OS varied significantly: 106.2 months for ISS I, and 79.1 and 63.1 months for ISS II and ISS III, respectively (p < 0.001). First-line treatment for 90% (n=1407) of patients included chemotherapy, with 70% of patients treated with Bortezomib-based triplets, with VCD (Bortezomib, Cyclophosphamide, and Dexamethasone) being the most frequently used of them (30.7%, n=486). Only 21.7% of patients were consolidated with autologous stem cell transplantation (ASCT) in the first line. Cohort A comprised 341 (21.6%) patients who died within 5 years after diagnosis. Univariate analysis showed a significantly higher proportion of patients ≥65 years, from the subsidized health plan, with renal insufficiency requiring dialysis, elevated B2-microglobulin levels at diagnosis, and who did not receive ASCT in this cohort. After multivariate logistic regression, four variables remained statistically significant predictors of OS: age ≥65 years, elevated B2-microglobulin levels, not receiving ASCT, and higher albumin values at diagnosis. These variables were associated with higher mortality risk, except for higher albumin values, which were associated with better survival (Table I). Conclusions: This study contributes to the growing body of evidence on multiple myeloma prognosis in the Colombian context. These findings provide valuable insights for clinicians in the local setting and may aid them in making informed decisions for individual patients. Nevertheless, future research and validation studies are warranted to confirm and build upon these findings.

Secure transmission of medical images in multi-cloud e-healthcare applications using data hiding scheme

In recent years, medical image transmission using a multi-cloud system has played a significant role in e-Healthcare infrastructure. It allows medical practitioners to easily store, retrieve, and share patients’ medical information across multiple stakeholders. However, multi-cloud image transmission may be vulnerable to multiple security breaches, such as authentication, confidentiality, and security issues. Motivated by these issues, this paper proposes a data-hiding scheme for secure medical image transmission in a multi-cloud environment. The proposed scheme ensures imperceptible robustness and watermark security at a low computational cost. Here, the medical image is divided into a number of shares using Neighbor Mean Interpolation (NMI). To achieve confidentiality, Electronic Patient Healthcare Record (EPHR) is encrypted using Double Scan Pixel Position Shuffling (DSPPS) approach. Then, the encrypted EPHR is divided into shares and embedded in the cover medical image shares. Finally, a minimum of 50% of watermarked image shares are utilized to retrieve the original medical image and encrypted EPHR, consequently reducing multi-cloud latency and computational burden. Experimental results show that the proposed scheme shows high imperceptibility, robustness, and watermark security at a low computational cost. Comparative analysis with some of the recent popular data hiding schemes shows that the proposed scheme has improved imperceptibility and robustness by 10%–15% (approximately) with higher watermark security at a low computational cost.

Assessing, updating and utilising primary care smoking records for lung cancer screening

BackgroundLung cancer screening with low-dose computed tomography for high-risk populations is being implemented in the UK. However, inclusive identification and invitation of the high-risk population is a major challenge for equitable lung screening implementation. Primary care electronic health records (EHRs) can be used to identify lung screening-eligible individuals based on age and smoking history, but the quality of EHR smoking data is limited. This study piloted a novel strategy for ascertaining smoking status in primary care and tested EHR search combinations to identify those potentially eligible for lung cancer screening.MethodsSeven primary care General Practices in South Wales, UK were included. Practice-level data on missing tobacco codes in EHRs were obtained. To update patient EHRs with no tobacco code, we developed and tested an algorithm that sent a text message request to patients via their GP practice to update their smoking status. The patient’s response automatically updated their EHR with the relevant tobacco code. Four search strategies using different combinations of tobacco codes for the age range 55-74+ 364 were tested to estimate the likely impact on the potential lung screening-eligible population in Wales. Search strategies included: BROAD (wide range of ever smoking codes); VOLUME (wide range of ever-smoking codes excluding “trivial” former smoking); FOCUSED (cigarette-related tobacco codes only), and RECENT (current smoking within the last 20 years).ResultsTobacco codes were not recorded for 3.3% of patients (n = 724/21,956). Of those with no tobacco code and a validated mobile telephone number (n = 333), 55% (n = 183) responded via text message with their smoking status. Of the 183 patients who responded, 43.2% (n = 79) had a history of smoking and were potentially eligible for lung cancer screening. Applying the BROAD search strategy was projected to result in an additional 148,522 patients eligible to receive an invitation for lung cancer screening when compared to the RECENT strategy.ConclusionAn automated text message system could be used to improve the completeness of primary care EHR smoking data in preparation for rolling out a national lung cancer screening programme. Varying the search strategy for tobacco codes may have profound implications for the size of the population eligible for lung-screening invitation.

Abstract 17709: ICD-10 Codes Have Low Sensitivity and Suboptimal Positive Predictive Value for Identifying Cases of Pulmonary Embolism With Cor Pulmonale in Electronic Healthcare Records

Introduction: The International Classification of Diseases, 10 th revision (ICD-10) codes were updated in 2015 to include designation of pulmonary embolism (PE) with cor pulmonale . These codes can be advantageous to identify higher-risk patients with PE for research or quality improvement. However, the validity of these codes is unknown. Methods: We reviewed patient data from Mass General Brigham (MGB) health system (01/2016-12/2021). Patients were randomly selected from 3 groups: Patients with Principal discharge diagnosis codes for PE, patients with secondary discharge diagnosis of PE, and patients with no PE codes. Accuracy of codes for PE with cor pulmonale were assessed with reference to manual review by 2 independent physicians using pre-specified criteria. Weighted estimates were obtained by considering the total number of hospitalizations. Results: 1,712 patients were included (age: 60.6±17.8 years, 52.3% female). Chart review found 865 patients with PE, of whom 360 (41.6%) had PE with cor pulmonale according to predefined clinical criteria. Of these 360 PE patients, 358 (99.4%) had ICD-10 codes for PE. However, only in 36 (10.1%) cases, the specific cor pulmonale codes were used. There were 24 additional cases with ICD-10 codes for PE with cor pulmonale in whom the diagnosis was not confirmed per manual review. Sensitivity and specificity for PE with cor pulmonale codes in Principal-or-secondary position were 10.0% and 98.2% (Table, panel A). The positive predictive value (PPV) for PE with cor pulmonale in Principal-or-secondary discharge position was 63.2% in the weighted sample (Table, panel B). Conclusions: ICD-10 codes for identification of acute PE with cor pulmonale have excellent specificity, but very poor sensitivity and suboptimal PPV. These results are informative for investigators who may use such ICD-10 codes for research or quality improvement purposes.

Abstract 17845: Logic-Based Natural Language Processing Identifies VT Recurrence in the Electronic Health Record

Introduction: With the advent of integrated electronic healthcare record (EHR) systems, it is increasingly important to identify patients who may suffer a clinical endpoint to allocate resources and develop predictive models. For the endpoint of VT recurrence after ablation, outcomes are currently assessed manually by experts, which is time consuming with few automated tools. We set out to develop a robust natural language processing model to parse cardiology and electrophysiology notes to detect VT. Hypothesis: We hypothesize that natural language processing (NLP) with logic-based algorithms can be used to accurately extract specific information on VT recurrence from complex clinical notes, compared to experts. Methods: We studied N=639 full-text clinical notes written by board certified cardiologists from N = 125 patients (32.0% female, LVEF 48.9±13.9%, 61±14.0 years). Notes were labeled as “No” or “Yes” for VT recurrence, and split into a development set of N=303 (median word count = 236) from N=75 patients and a test set of N=336 (median words = 297) from N=50 patients. We developed an NLP rule based expert system that uses hierarchical groups of keywords to identify sentences related to VT, then uses negation and quantifier functions to interpret context and assign a label. Results: The logical flow is shown in figure A. This approach achieved an 93.73% accuracy in the development set (precision = 57.14%, recall = 69.56%), and 90.48% accuracy (precision = 36.84%, recall = 63.63%) in the test set (figure B). Conclusions: An NLP logic-based approach can identify VT recurrence post ablation from complex clinical notes. NLP can thus be used to accelerate data extraction for clinical purposes or for semi-supervised learning. Future work will determine if logic-based NLP need to be augmented for noisier data, such as the entire EHR system.

Diagnostic accuracy, clinical characteristics, and prognostic differences of patients with acute myocarditis according to inclusion criteria.

The diagnosis of acute myocarditis (AM) is complex due to its heterogeneity and typically is defined by either Electronic Healthcare Records (EHRs) or advanced imaging and endomyocardial biopsy, but there is no consensus. We aimed to investigate the diagnostic accuracy of these approaches for AM. Data on ICD 10th Revision(ICD-10) codes corresponding to AM were collected from two hospitals and compared to cardiac magnetic resonance (CMR)-confirmed or clinically suspected (CS)-AM cases with respect to diagnostic accuracy, clinical characteristics, and all-cause mortality. Next, we performed a review of published AM studies according to inclusion criteria. We identified 291 unique admissions with ICD-10 codes corresponding to AM in the first three diagnostic positions. The positive predictive value of ICD-10 codes for CMR-confirmed or CS-AM was 36%, and patients with CMR-confirmed or CS-AM had a lower all-cause mortality than those with a refuted diagnosis (P=0.019). Using an unstructured approach, patients with CMR-confirmed and CS-AM had similar demographics, comorbidity profiles and survival over a median follow-up of 52 months (P=0.72). Our review of the literature confirmed our findings. Outcomes for patients included in studies using CMR-confirmed criteria were favourable compared to studies with endomyocardial biopsy-confirmed AM cases. ICD-10 codes have poor accuracy in identification of AM cases and should be used with caution in clinical research. There are important differences in management and outcomes of patients according to the selection criteria used to diagnose AM. Potential selection biases must be considered when interpreting AM cohorts and requires standardization of inclusion criteria for AM studies.

Comparison Between Pantoprazole Intermittent Dosing and Continuous Infusion in Suspected Upper Gastrointestinal Bleeding Prior to Endoscopy: Impact of a Pharmacist-Driven Protocol to Reduce Utilization of Pantoprazole Continuous Infusion.

Current practice for patients with suspected or confirmed upper gastrointestinal bleeding (GIB) is to utilize a proton pump inhibitor (PPI) bolus followed by a continuous infusion for 72 hours. Literature has shown similar outcomes with intermittent bolus dosing compared to continuous infusion. Substitution would lead to reduced costs and utilization of resources. This was a retrospective case-control study conducted via chart review. Utilizing electronic healthcare record reports, patients in the control arm were screened for inclusion if they received a pantoprazole continuous infusion from December 1, 2020, to March 31, 2021. A total of 38 patients were included in the control arm. Patients in the experimental arm were screened for inclusion with pantoprazole intermittent therapy from January 1, 2022, to June 30, 2022. A total of 60 patients were included in the experimental arm. The primary outcome was a 30-day GIB recurrence. Secondary outcomes included 30-day hospital readmission, 30-day Clostridioides difficile (C. difficile), hospital length of stay (LOS), and number of pantoprazole vials utilized. There was a 65% reduction in the 30-day GIB recurrence in the intermittent bolus arm compared to the continuous infusion arm. Thirty-day hospital readmission was 57% lower in the intermittent bolus arm compared to the continuous infusion arm. The LOS between the two arms was almost identical with the median being five days for the intermittent bolus arm and the median being four days for the continuous infusion arm. The 30-day C. difficile infection rate had 5% of patients acquiring C. difficile in the intermittent bolus arm and 2.5% in the continuous infusion arm. The intermittent bolus arm used 55% fewer pantoprazole vials than the continuous infusion arm. In hospitalized patients, the utilization of pantoprazole intermittent bolus is not only comparably efficacious but potentially represents a safer and economically advantageous alternative compared to the current guideline recommendation of a 72-hour pantoprazole continuous infusion. Further studies could provide more robust data to support our findings and challenge the current recommendation for patients who meet the indication criteria.

Developed A Hybrid Optimal Feature Vector Selection with Blockchain Technology for Smart Healthcare 4.0

The health economy has been an innovative technology since time-honored. Preserving and maintaining patient data are essential in a routine life. Patient’s medical information is very important for every individual not only for patients but also for doctors who are examining them. Advances in sensing technology, processing of data, and communication protocols have transformed the healthcare industry. Patients, physicians, hospitals, and other stakeholder may keep vital data and medical records with the use of electronic healthcare records (EHR). The goal of research should be to develop a Hybrid Optimal Feature Vector Selection with Blockchain Technology (HOFVS-BT) for smart healthcare 4.0 to improve the secure transmission of data which is supported intelligent IoT and medical detection platform possible. For Feature vector selection, proposed an Orthogonal Wolf Optimization (OWO) algorithm. Furthermore, safeguarding private patient details is taken into account by establishing an upgraded Blockchain-based IoT data security solution that not only secures the data, but also fosters trust between patients/users and healthcare service providers.

Prediction models for in-hospital deaths of patients with COVID-19 using electronic healthcare data

Objective Many models for predicting various disease prognoses have achieved high performance without laboratory test results. However, whether laboratory test results can improve performance remains unclear. This study aimed to investigate whether laboratory test results improve the model performance for coronavirus disease 2019 (COVID-19). Methods Prediction models were developed using data from the electronic healthcare record database in Japan. Patients aged ≥18 years hospitalized for COVID-19 after February 11, 2020, were included. Their age, sex, comorbidities, laboratory test results, and number of days from February 11, 2020, were collected. We developed a logistic regression, XGBOOST, random forest, and neural network analysis and compared the performance with and without laboratory test results. The performance of predicting in-hospital death was evaluated using the area under the curve (AUC). Results Data from 8,288 hospitalized patients (females, 46.5%) were analyzed. The median patient age was 71 years. A total of 6,630 patients were included in the training dataset, and 312 (4.7%) died. In the logistic regression model, the area under the curve was 0.88 (95% confidence interval [CI] = 0.83–0.93) and 0.75 (95% CI = 0.68–0.81) with and without laboratory test results, respectively. The performance was not fundamentally different between the model types, and the laboratory test results improved the performance in all cases. The variables useful for prediction were blood urea nitrogen, albumin, and lactate dehydrogenase. Conclusions Laboratory test results, such as blood urea nitrogen, albumin, and lactate dehydrogenase levels, along with background information, helped estimate the prognosis of patients hospitalized for COVID-19.

Dynamic relationship among immediate release fentanyl use and cancer incidence: A multivariate time-series analysis using vector autoregressive models

Background A substantial increase in the incidence of immediate release fentanyl (IRF) use was reported in Spain from 2012 to 2017. Purpose This study aimed to investigate the relationship dynamically with cancer incidence in order to provide empirical evidence of inappropriate use of IRF with respect to the pathology. Research design A vector autoregresive (VAR) model was constructed using data from a nationwide electronic healthcare record database in primary care in Spain (BIFAP) according to the following step procedure: (1) split data into training data for modelling and test for validation (2) assessing for time series stationarity; (3) selecting lag-length; (4) building the VAR model; (5) assessing residual autocorrelation; (6) checking stability of the VAR system; (7) evaluating Granger causality; (8) impulse response analysis and forecast error variance decomposition (9) prediction performance with validation data. Results The analysis showed a strong and linear correlation between IRF and cancer (Pearson correlation coefficient: 0.594 (95% CI: 0.420–0.726). Two VAR models, VAR (2) and VAR (11) were selected and compared. All tests performed for both models satisfied assumptions for stability, predictability and accuracy. Granger causality revealed cancer incidence is a good predictor for IRF use. VAR (2) seemed to be slightly more accurate, according to the RMSE of the test data. Conclusions This study demonstrates that using a robust and structured VAR modelling approach, is able to estimate dynamics associations, involving IRF use and cancer incidence.

Impacts of an Electronic Health Record Transition on Veterans Health Administration Health Professions Trainee Experience.

Adoption of electronic health care records (EHRs) has proliferated since 2000. While EHR transitions are widely understood to be disruptive, little attention has been paid to their effect on health professions trainees' (HPTs) ability to learn and conduct work. Veterans Health Administration's (VA) massive transition from its homegrown EHR (CPRS/Vista) to the commercial Oracle Cerner presents an unparalleled-in-scope opportunity to gain insight on trainee work functions and their ability to obtain requisite experience during transitions. To identify how an organizational EHR transition affected HPT work and learning at the third VA go-live site. A formative mixed-method evaluation of HPT experiences with VHA's EHR transition including interviews with HPTs and supervisors at Chalmers P. Wylie VA Outpatient Clinic in Columbus, OH, before (~60 min), during (15-30 min), and after (~60 min) go-live (December 2021-July 2022). We also conducted pre- (March 2022-April 2022) and post-go live (May 2022-June 2022) HPT and employee surveys. We conducted 24 interviews with HPTs (n=4), site leaders (n=2), and academic affiliates (n=2) using snowball sampling. We recruited HPTs in pre- (n=13) and post-go-live (n=10) surveys and employees in pre- (n=408) and post-go-live (n=458) surveys. We conducted interviews using a semi-structured guide and grounded prompts. We coded interviews and survey free text data using a priori and emergent codes, subsequently conducting thematic analysis. We conducted descriptive statistical analysis of survey responses and merged interview and survey data streams. Our preliminary findings indicate that the EHR transition comprehensively affected HPT experiences, disrupting processes from onboarding and training to clinical care contributions and training-to-career retention. Understanding HPTs' challenges during EHR transitions is critical to effective training. Mitigating the identified barriers to HPT training and providing patient care may lessen their dissatisfaction and ensure quality patient care during EHR transitions.

Development of an Asthma Exacerbation Risk Prediction Model for Conversational Use by Adults in England.

Improving accurate risk assessment of asthma exacerbations, and reduction via relevant behaviour change among people with asthma could save lives and reduce health care costs. We developed a simple personalised risk prediction model for asthma exacerbations using factors collected in routine healthcare data for use in a risk modelling feature for automated conversational systems. We used pseudonymised primary care electronic healthcare records from the Clinical Practice Research Datalink (CPRD) Aurum database in England. We combined variables for prediction of asthma exacerbations using logistic regression including age, gender, ethnicity, Index of Multiple Deprivation, geographical region and clinical variables related to asthma events. We included 1,203,741 patients divided into three cohorts to implement temporal validation: 898,763 (74.7%) in the training sample, 226,754 (18.8%) in the testing sample and 78,224 (6.5%) in the validation sample. The Area under the ROC curve (AUC) for the full model was 0.72 and for the restricted model was 0.71. Using a cut-off point of 0.1, approximately 27 asthma reviews by clinicians per 100 patients would be prevented compared with a strategy that all patients are regarded as high risk. Compared with patients without an exacerbation, patients who exacerbated were older, more likely to be female, prescribed more SABA and ICS in the preceding 12 months, have history of GORD, COPD, anxiety, depression, live in very deprived areas and have more severe disease. Using information available from routinely collected electronic healthcare record data, we developed a model that has moderate ability to separate patients who had an asthma exacerbation within 3 months from their index date from patients who did not. When comparing this model with a simplified model with variables that can easily be self-reported through a WhatsApp chatbot, we have shown that the predictive performance of the model is not substantially different.

Health chain-block chain based electronic healthcare record system with access and permission management

An electronic health record (EHR) is a technology that allows you to keep track of your health information. It keeps computerized records of several healthcare organizations. Records are exchanged via enterprise-wide data systems as well as other networking technologies and exchanges. Patients nowadays expect immediate access to their health information. However, the health sector comes with immediate access to data, and there are worries about the privacy and security of medical records of patients. As a result, a blockchain-based solution can assist in resolving this issue. The blockchain has the potential to beat the conventional centralized system, which suffers from a severe lack of accessibility. The current healthcare system is often seen as being both costly and convoluted. However, with improved health data management and insurance providers, this may be reduced via the use of Blockchain Technology (BT). It is the goal of the developers of BT to make it possible to keep track of financial transactions without having to rely on a central authority or a bank. Innovations in BT have allowed for better health records, insurance billing, and smart contract transactions, and it also enable a decentralized data structure of activities, as well as eternal availability and security over health information. Access to medical records, instrument tracking, prescription databases, hospital facilities, and property lifetime all inside a blockchain architecture is made possible by BT that also greatly improves the seamless integration of health information. The use of BT in the medical field has this advantage as one of its primary advantages. Access to a patient's medical history is crucial for prescribing the right treatment, and BT has the potential to greatly enhance the current state of public healthcare delivery. This paper offers a block chain-based scheme called “Health Chain” as a solution for maintaining medical records. Health Chain is made to be better than the existing systems since it gives patients, health professionals, and other external stakeholders safe, efficient, and ubiquitous medical information available while maintaining the patients' confidentiality. To regulate operations and restrict access to digital health records, Health Chain uses timed-based smart contracts. In addition, it uses cutting-edge encryption methods to boost security. This study suggests a new reward system that uses healthcare professionals to track patients' medical information and construct additional units. Numerous tests are run to assess the proposed Health Chain model's performance, and the findings demonstrate that the proposed suggestion effectively manages a huge dataset called (nCOV19) from the Covid-19 study with little delay. Maximum efficiency will be expected to be attained by 98%, transaction latency by 95%, overall system execution time by 96%, data security by 92%, and data scalability by 95%, according to the suggested research.

Development and feasibility of a telemedicine tool for patients with recurrent urinary tract infection: myRUTIcoach

Introduction and hypothesisPatients with recurrent urinary tract infection (rUTI) have limited knowledge of preventive strategies to lower the risk of UTI. We aimed to develop and test the feasibility of an eHealth system for women with rUTI, named myRUTIcoach, and explored the facilitators and barriers related to its adoption.MethodsWe developed myRUTIcoach in a structured iterative process and tested its feasibility among 25 women with rUTI over 2 months. Subsequent questionnaires covered satisfaction, accessibility, and experiences with myRUTIcoach. A random selection of participants and relevant stakeholders took part in semi-structured interviews to explore adoption. Data were analyzed and elaborated using inductive and deductive approaches using the Non-adoption, Abandonment, Spread, Scale-up, and Sustainability (NASSS) framework.Results MyRUTIcoach was not only widely accepted but also facilitated communication with health care professionals (HCPs) and contributed to greater knowledge of rUTI. Women graded the system a mean of 8.0 (±0.6) out of 10, with 89% stating that they would recommend it to others. Patients indicated that self-management skills were the major facilitators and barriers related to adoption, whereas HCPs stated that the disconnect between myRUTIcoach and electronic health care records (EHRs) was the major barrier.ConclusionsThis research describes the development and testing of myRUTIcoach for women with rUTI. Patients and HCPs reported high satisfaction and compliance with myRUTIcoach. However, adoption by the intended users is complex and influenced by all examined domains of the NASSS framework. We have already improved linkage to EHRs, but further optimization to meet patient needs may improve the effectiveness of this self-management tool for rUTI.

Augmented intelligence facilitates concept mapping across different electronic health records

IntroductionWith the advent of artificial intelligence, the secondary use of routinely collected medical data from electronic healthcare records (EHR) has become increasingly popular. However, different EHR systems typically use different names for the same medical concepts. This obviously hampers scalable model development and subsequent clinical implementation for decision support. Therefore, converting original parameter names to a so-called ontology, a standardized set of predefined concepts, is necessary but time-consuming and labor-intensive. We therefore propose an augmented intelligence approach to facilitate ontology alignment by predicting correct concepts based on parameter names from raw electronic health record data exports. MethodsWe used the manually mapped parameter names from the multicenter “Dutch ICU data warehouse against COVID-19” sourced from three types of EHR systems to train machine learning models for concept mapping. Data from 29 intensive care units on 38,824 parameters mapped to 1,679 relevant and unique concepts and 38,069 parameters labeled as irrelevant were used for model development and validation. We used the Natural Language Toolkit (NLTK) to preprocess the parameter names based on WordNet cognitive synonyms transformed by term-frequency inverse document frequency (TF-IDF), yielding numeric features. We then trained linear classifiers using stochastic gradient descent for multi-class prediction. Finally, we fine-tuned these predictions using information on distributions of the data associated with each parameter name through similarity score and skewness comparisons. ResultsThe initial model, trained using data from one hospital organization for each of three EHR systems, scored an overall top 1 precision of 0.744, recall of 0.771, and F1-score of 0.737 on a total of 58,804 parameters. Leave-one-hospital-out analysis returned an average top 1 recall of 0.680 for relevant parameters, which increased to 0.905 for the top 5 predictions. When reducing the training dataset to only include relevant parameters, top 1 recall was 0.811 and top 5 recall was 0.914 for relevant parameters. Performance improvement based on similarity score or skewness comparisons affected at most 5.23% of numeric parameters. ConclusionAugmented intelligence is a promising method to improve concept mapping of parameter names from raw electronic health record data exports. We propose a robust method for mapping data across various domains, facilitating the integration of diverse data sources. However, recall is not perfect, and therefore manual validation of mapping remains essential.

Validation of a dynamic method of measuring households and populations from primary care Electronic Health Records: Cross-sectional comparison with Office for National Statistics Census 2021 estimates

ObjectivesWe have developed a dynamic method for identifying household members from Electronic Health Records (EHR). We compared the 2021 Census estimates of household number and demography with similar estimates derived from primary care EHRs on the Census date using primary care EHRs for the population of north east London (NEL).
 MethodWe included 2,115,017 patients registered with a general practitioner on the 2021 Census date in NEL and assigned households from encrypted Unique Property Reference Numbers. We compared household number and size by Local Authority (LA), Middle Layer Super Output Area (MSOA) and area’s Index of Multiple Deprivation quintiles (IMDq) to Office for National Statistics (ONS) 2021 Census estimates and by LA to ONS Admin Based Housing Stock (ABHS) 2020 estimates. We assessed differences in EHR and Census 2021 populations by sex, age, LA, MSOA and IMDq. Sensitivity analyses will exclude those without a recent recorded clinical encounter.
 ResultsEHR population estimates (2,115,017) were 116,346 (5.8%) higher than Census estimates (1,998,671), higher among men (9.2%) than women (2.5%) in almost all age groups, especially men aged 30-50 years and higher in the most (8.7%), than in the least (2.5%) deprived IMDq. EHR household estimates (660,789) were 68,047 (9.3%) lower than Census estimates (728,836), and 19,719 (3.1%) higher than ABHS occupied addresses (641,070). EHR household size estimates were 15.6%, 29.2%,12.5% and 8.4% lower for household sizes 1,2 3 and 4, and 13.3%, 42.1%, 82.1% and 195.8% higher for household sizes 5, 6,7 and 8 respectively when compared to Census estimates. EHR population and household estimates were respectively 5-10% higher and 5-11% lower for almost all NEL local authorities.
 ConclusionEHR- and Census-derived population and household estimates differ, mainly in the prevalence of larger households. While data were extracted on the same date, person-level validation was not possible. Differences may reflect deregistration delay in EHR when changing residence. Analyses based on clinical encounters recency may identify registered patients who are no longer residents.