Possible Sources of Bias in Primary Care Electronic Health Record Data Use and Reuse

Development and initial validation of a data quality evaluation tool in obstetrics real-world data through HL7-FHIR interoperable Bayesian networks and expert rules

Background

The increasing prevalence of electronic health records (EHRs) in healthcare systems globally has underscored the importance of data quality for clinical decision-making and research, particularly in obstetrics. High-quality data is vital for an accurate representation of patient populations and to avoid erroneous healthcare decisions. However, existing studies have highlighted significant challenges in EHR data quality, necessitating innovative tools and methodologies for effective data quality assessment and improvement.

Objective

This article addresses the critical need for data quality evaluation in obstetrics by developing a novel tool. The tool utilizes Health Level 7 (HL7) Fast Healthcare Interoperable Resources (FHIR) standards in conjunction with Bayesian Networks and expert rules, offering a novel approach to assessing data quality in real-world obstetrics data.

Methods

A harmonized framework focusing on completeness, plausibility, and conformance underpins our methodology. We employed Bayesian networks for advanced probabilistic modeling, integrated outlier detection methods, and a rule-based system grounded in domain-specific knowledge. The development and validation of the tool were based on obstetrics data from 9 Portuguese hospitals, spanning the years 2019-2020.

Results

The developed tool demonstrated strong potential for identifying data quality issues in obstetrics EHRs. Bayesian networks used in the tool showed high performance for various features with area under the receiver operating characteristic curve (AUROC) between 75% and 97%. The tool's infrastructure and interoperable format as a FHIR Application Programming Interface (API) enables a possible deployment of a real-time data quality assessment in obstetrics settings. Our initial assessments show promised, even when compared with physicians' assessment of real records, the tool can reach AUROC of 88%, depending on the threshold defined.

Discussion

Our results also show that obstetrics clinical records are difficult to assess in terms of quality and assessments like ours could benefit from more categorical approaches of ranking between bad and good quality.

Conclusion

This study contributes significantly to the field of EHR data quality assessment, with a specific focus on obstetrics. The combination of HL7-FHIR interoperability, machine learning techniques, and expert knowledge presents a robust, adaptable solution to the challenges of healthcare data quality. Future research should explore tailored data quality evaluations for different healthcare contexts, as well as further validation of the tool capabilities, enhancing the tool's utility across diverse medical domains.

Using Electronic Health Records to Facilitate Precision Psychiatry

The use of clinical prediction models to produce individualized risk estimates can facilitate the implementation of precision psychiatry. As a source of data from large, clinically representative patient samples, electronic health records (EHRs) provide a platform to develop and validate clinical prediction models, as well as potentially implement them in routine clinical care. The current review describes promising use cases for the application of precision psychiatry to EHR data and considers their performance in terms of discrimination (ability to separate individuals with and without the outcome) and calibration (extent to which predicted risk estimates correspond to observed outcomes), as well as their potential clinical utility (weighing benefits and costs associated with the model compared to different approaches across different assumptions of the number needed to test). We review 4 externally validated clinical prediction models designed to predict psychosis onset, psychotic relapse, cardiometabolic morbidity, and suicide risk. We then discuss the prospects for clinically implementing these models and the potential added value of integrating data from evidence syntheses, standardized psychometric assessments, and biological data into EHRs. Clinical prediction models can utilize routinely collected EHR data in an innovative way, representing a unique opportunity to inform real-world clinical decision making. Combining data from other sources (e.g., meta-analyses) or enhancing EHR data with information from research studies (clinical and biomarker data) may enhance our abilities to improve the performance of clinical prediction models.

Impact of the COVID-19 Pandemic on Foodborne Disease Healthcare-Seeking Behavior and Diagnoses at a Large Academic Medical System

The objective of this study was to examine changes in healthcare-seeking behaviors and diagnostic practices around foodborne illness during the COVID-19 pandemic in a large university-based health system. A retrospective cohort study of individuals diagnosed with pathogens commonly transmitted through food between 2015 and 2020 was undertaken using electronic medical record data. Regression models were used to compare measured incidence rates of various foodborne pathogens as well as associated healthcare-seeking behaviors during the pandemic year of 2020 to previous years. Incidence of campylobacteriosis, cholera, and norovirus in 2020 significantly decreased, respectively, by 65.5% (p < 0.01), 90.1% (p = 0.02), and 73.0% (p = 0.03) compared with an average from 2017- to 019. Average annual visits for patients included in our sample significantly increased by 8.0% when comparing the average from 2017-2019 to 2020 (p < 0.01). These results suggest that the pandemic impacted healthcare use related to foodborne disease either due to reduced exposure to foodborne pathogens or reduced willingness to seek healthcare.

Predicting unplanned admissions to hospital in older adults using routinely recorded general practice data: development and validation of a prediction model

BACKGROUND

Unplanned admissions to hospital represent a hazardous event for older people. Timely identification of high-risk individuals using a prediction tool may facilitate preventive interventions.

AIM

To develop and validate an easy-to-use prediction model for unplanned admissions to hospital in community-dwelling older adults using readily available data to allow rapid bedside assessment by GPs.

DESIGN AND SETTING

This was a retrospective study using the general practice electronic health records of 243 324 community-dwelling adults aged ≥65 years linked with national administrative data to predict unplanned admissions to hospital within 6 months.

METHOD

The dataset was geographically split into a development (n = 142 791/243 324, 58.7%) and validation (n = 100 533/243 324, 41.3%) sample to predict unplanned admissions to hospital within 6 months. The performance of three different models was evaluated with increasingly smaller selections of candidate predictors (optimal, readily available, and easy-to-use models). Logistic regression was used with backward selection for model development. The models were validated internally and externally. Predictive performance was assessed by area under the curve (AUC) and calibration plots.

RESULTS

In both samples, 7.6% (development cohort: n = 10 839/142 791, validation cohort: n = 7675/100 533) had ≥1 unplanned hospital admission within 6 months. The discriminative ability of the three models was comparable and remained stable after geographic validation. The easy-to-use model included age, sex, prior admissions to hospital, pulmonary emphysema, heart failure, and polypharmacy. Its discriminative ability after validation was AUC 0.72 (95% confidence interval = 0.71 to 0.72). Calibration plots showed good calibration.

CONCLUSION

The models showed satisfactory predictive ability. Reducing the number of predictors and geographic validation did not have an impact on predictive performance, demonstrating the robustness of the model. An easy-to-use tool has been developed in this study that may assist GPs in decision making and with targeted preventive interventions.

From admission to discharge: a systematic review of clinical natural language processing along the patient journey

BACKGROUND

Medical text, as part of an electronic health record, is an essential information source in healthcare. Although natural language processing (NLP) techniques for medical text are developing fast, successful transfer into clinical practice has been rare. Especially the hospital domain offers great potential while facing several challenges including many documents per patient, multiple departments and complex interrelated processes.

METHODS

In this work, we survey relevant literature to identify and classify approaches which exploit NLP in the clinical context. Our contribution involves a systematic mapping of related research onto a prototypical patient journey in the hospital, along which medical documents are created, processed and consumed by hospital staff and patients themselves. Specifically, we reviewed which dataset types, dataset languages, model architectures and tasks are researched in current clinical NLP research. Additionally, we extract and analyze major obstacles during development and implementation. We discuss options to address them and argue for a focus on bias mitigation and model explainability.

RESULTS

While a patient's hospital journey produces a significant amount of structured and unstructured documents, certain steps and documents receive more research attention than others. Diagnosis, Admission and Discharge are clinical patient steps that are researched often across the surveyed paper. In contrast, our findings reveal significant under-researched areas such as Treatment, Billing, After Care, and Smart Home. Leveraging NLP in these stages can greatly enhance clinical decision-making and patient outcomes. Additionally, clinical NLP models are mostly based on radiology reports, discharge letters and admission notes, even though we have shown that many other documents are produced throughout the patient journey. There is a significant opportunity in analyzing a wider range of medical documents produced throughout the patient journey to improve the applicability and impact of NLP in healthcare.

CONCLUSIONS

Our findings suggest that there is a significant opportunity to leverage NLP approaches to advance clinical decision-making systems, as there remains a considerable understudied potential for the analysis of patient journey data.

Exploring the barriers, facilitators and needs to use patient outcomes in district nursing care: A multi-method qualitative study

AIM AND OBJECTIVES

To provide an in-depth insight into the barriers, facilitators and needs of district nurses and nurse assistants on using patient outcomes in district nursing care.

BACKGROUND

As healthcare demands grow, particularly in district nursing, there is a significant need to understand how to systematically measure and improve patient outcomes in this setting. Further investigation is needed to identify the barriers and facilitators for effective implementation.

DESIGN

A multi-method qualitative study.

METHODS

Open-ended questions of a survey study (N = 132) were supplemented with in-depth online focus group interviews involving district nurses and nurse assistants (N = 26) in the Netherlands. Data were analysed using thematic analysis.

RESULTS

Different barriers, facilitators and needs were identified and compiled into 16 preconditions for using outcomes in district nursing care. These preconditions were summarised into six overarching themes: follow the steps of a learning healthcare system; provide patient-centred care; promote the professional's autonomy, attitude, knowledge and skills; enhance shared responsibility and collaborations within and outside organisational boundaries; prioritise and invest in the use of outcomes; and boost the unity and appreciation for district nursing care.

CONCLUSIONS

The preconditions identified in this study are crucial for nurses, care providers, policymakers and payers in implementing the use of patient outcomes in district nursing practice. Further exploration of appropriate strategies is necessary for a successful implementation.

RELEVANCE TO CLINICAL PRACTICE

This study represents a significant step towards implementing the use of patient outcomes in district nursing care. While most research has focused on hospitals and general practitioner settings, this study focuses on the needs for district nursing care. By identifying 16 key preconditions across themes such as patient-centred care, professional autonomy and unity, the findings offer valuable guidance for integrating a learning healthcare system that prioritises the measurement and continuous improvement of patient outcomes in district nursing.

REPORTING METHOD

Consolidated Criteria for Reporting Qualitative Research (COREQ) guidelines.

PATIENT OF PUBLIC CONTRIBUTION

No patient or public contribution.

Supplementing systematic review findings with healthcare system data: pilot projects from the Agency for Healthcare Research and Quality Evidence-based Practice Center program

OBJECTIVES

The US Agency for Healthcare Research and Quality, through the Evidence-based Practice Center (EPC) Program, aims to provide health system decision makers with the highest-quality evidence to inform clinical decisions. However, limitations in the literature may lead to inconclusive findings in EPC systematic reviews (SRs). The EPC Program conducted pilot projects to understand the feasibility, benefits, and challenges of utilizing health system data to augment SR findings to support confidence in healthcare decision-making based on real-world experiences.

STUDY DESIGN AND SETTING

Three contractors (each an EPC located at a different health system) selected a recently completed SR conducted by their center and identified an evidence gap that electronic health record (EHR) data might address. All pilot project topics addressed clinical questions as opposed to care delivery, care organization, or care disparities topics that are common in EPC reports. Topic areas addressed by each EPC included infantile epilepsy, migraine, and hip fracture. EPCs also tracked additional resources needed to conduct supplemental analyses. The workgroup met monthly in 2022-2023 to discuss challenges and lessons learned from the pilot projects.

RESULTS

Two supplemental data analyses filled an evidence gap identified in the SRs (raised certainty of evidence, improved applicability) and the third filled a health system knowledge gap. Project challenges fell under three themes: regulatory and logistical issues, data collection and analysis, and interpretation and presentation of findings. Limited ability to capture key clinical variables given inconsistent or missing data within the EHR was a major limitation. The workgroup found that conducting supplemental data analysis alongside an SR was feasible but adds considerable time and resources to the review process (estimated total hours to complete pilot projects ranged from 283 to 595 across EPCs), and that the increased effort and resources added limited incremental value.

CONCLUSION

Supplementing existing SRs with analyses of EHR data is resource intensive and requires specialized skillsets throughout the process. While using EHR data for research has immense potential to generate real-world evidence and fill knowledge gaps, these data may not yet be ready for routine use alongside SRs.

Centralized and Federated Models for the Analysis of Clinical Data

The progress of precision medicine research hinges on the gathering and analysis of extensive and diverse clinical datasets. With the continued expansion of modalities, scales, and sources of clinical datasets, it becomes imperative to devise methods for aggregating information from these varied sources to achieve a comprehensive understanding of diseases. In this review, we describe two important approaches for the analysis of diverse clinical datasets, namely the centralized model and federated model. We compare and contrast the strengths and weaknesses inherent in each model and present recent progress in methodologies and their associated challenges. Finally, we present an outlook on the opportunities that both models hold for the future analysis of clinical data.

Predictive value of abnormal blood tests for detecting cancer in primary care patients with nonspecific abdominal symptoms: A population-based cohort study of 477,870 patients in England

BACKGROUND

Identifying patients presenting with nonspecific abdominal symptoms who have underlying cancer is a challenge. Common blood tests are widely used to investigate these symptoms in primary care, but their predictive value for detecting cancer in this context is unknown. We quantify the predictive value of 19 abnormal blood test results for detecting underlying cancer in patients presenting with 2 nonspecific abdominal symptoms.

METHODS AND FINDINGS

Using data from the UK Clinical Practice Research Datalink (CPRD) linked to the National Cancer Registry, Hospital Episode Statistics and Index of Multiple Deprivation, we conducted a population-based cohort study of patients aged ≥30 presenting to English general practice with abdominal pain or bloating between January 2007 and October 2016. Positive and negative predictive values (PPV and NPV), sensitivity, and specificity for cancer diagnosis (overall and by cancer site) were calculated for 19 abnormal blood test results co-occurring in primary care within 3 months of abdominal pain or bloating presentations. A total of 9,427/425,549 (2.2%) patients with abdominal pain and 1,148/52,321 (2.2%) with abdominal bloating were diagnosed with cancer within 12 months post-presentation. For both symptoms, in both males and females aged ≥60, the PPV for cancer exceeded the 3% risk threshold used by the UK National Institute for Health and Care Excellence for recommending urgent specialist cancer referral. Concurrent blood tests were performed in two thirds of all patients (64% with abdominal pain and 70% with bloating). In patients aged 30 to 59, several blood abnormalities updated a patient's cancer risk to above the 3% threshold: For example, in females aged 50 to 59 with abdominal bloating, pre-blood test cancer risk of 1.6% increased to: 10% with raised ferritin, 9% with low albumin, 8% with raised platelets, 6% with raised inflammatory markers, and 4% with anaemia. Compared to risk assessment solely based on presenting symptom, age and sex, for every 1,000 patients with abdominal bloating, assessment incorporating information from blood test results would result in 63 additional urgent suspected cancer referrals and would identify 3 extra cancer patients through this route (a 16% relative increase in cancer diagnosis yield). Study limitations include reliance on completeness of coding of symptoms in primary care records and possible variation in PPVs if extrapolated to healthcare settings with higher or lower rates of blood test use.

CONCLUSIONS

In patients consulting with nonspecific abdominal symptoms, the assessment of cancer risk based on symptoms, age and sex alone can be substantially enhanced by considering additional information from common blood test results. Male and female patients aged ≥60 presenting to primary care with abdominal pain or bloating warrant consideration for urgent cancer referral or investigation. Further cancer assessment should also be considered in patients aged 30 to 59 with concurrent blood test abnormalities. This approach can detect additional patients with underlying cancer through expedited referral routes and can guide decisions on specialist referrals and investigation strategies for different cancer sites.

Trends and Persistent Disparities in Child Obesity During the COVID-19 Pandemic

The COVID-19 pandemic has been associated with increases in pediatric obesity and widening pre-existing disparities. To better understand the pandemic's long-term impacts, we evaluated trends in obesity across different demographic groups during the pandemic through December 2022. Using a retrospective cohort design, we analyzed electronic health record data from a large pediatric primary care network. Logistic regression models fit using generalized estimating equations estimated odds ratios (ORs) for changes in the level and trajectory of obesity across 2-year month-matched periods: prepandemic (June 2017 to December 2019) and pandemic (June 2020 to December 2022). Among a cohort of 153,667 patients with visits in each period, there was a significant increase in the level of obesity at the pandemic onset [OR: 1.229, 95% confidence interval (CI): 1.211-1.247] followed by a significant decrease in the trend for obesity (OR: 0.993, 95% CI: 0.992-0.993). By December 2022, obesity had returned to prepandemic levels. However, persistent sociodemographic disparities remain.

Natural language processing systems for extracting information from electronic health records about activities of daily living. A systematic review

Objective

Natural language processing (NLP) can enhance research on activities of daily living (ADL) by extracting structured information from unstructured electronic health records (EHRs) notes. This review aims to give insight into the state-of-the-art, usability, and performance of NLP systems to extract information on ADL from EHRs.

Materials and Methods

A systematic review was conducted based on searches in Pubmed, Embase, Cinahl, Web of Science, and Scopus. Studies published between 2017 and 2022 were selected based on predefined eligibility criteria.

Results

The review identified 22 studies. Most studies (65%) used NLP for classifying unstructured EHR data on 1 or 2 ADL. Deep learning, combined with a ruled-based method or machine learning, was the approach most commonly used. NLP systems varied widely in terms of the pre-processing and algorithms. Common performance evaluation methods were cross-validation and train/test datasets, with F1, precision, and sensitivity as the most frequently reported evaluation metrics. Most studies reported relativity high overall scores on the evaluation metrics.

Discussion

NLP systems are valuable for the extraction of unstructured EHR data on ADL. However, comparing the performance of NLP systems is difficult due to the diversity of the studies and challenges related to the dataset, including restricted access to EHR data, inadequate documentation, lack of granularity, and small datasets.

Conclusion

This systematic review indicates that NLP is promising for deriving information on ADL from unstructured EHR notes. However, what the best-performing NLP system is, depends on characteristics of the dataset, research question, and type of ADL.

Comparing natural language processing representations of coded disease sequences for prediction in electronic health records

OBJECTIVE

Natural language processing (NLP) algorithms are increasingly being applied to obtain unsupervised representations of electronic health record (EHR) data, but their comparative performance at predicting clinical endpoints remains unclear. Our objective was to compare the performance of unsupervised representations of sequences of disease codes generated by bag-of-words versus sequence-based NLP algorithms at predicting clinically relevant outcomes.

MATERIALS AND METHODS

This cohort study used primary care EHRs from 6 286 233 people with Multiple Long-Term Conditions in England. For each patient, an unsupervised vector representation of their time-ordered sequences of diseases was generated using 2 input strategies (212 disease categories versus 9462 diagnostic codes) and different NLP algorithms (Latent Dirichlet Allocation, doc2vec, and 2 transformer models designed for EHRs). We also developed a transformer architecture, named EHR-BERT, incorporating sociodemographic information. We compared the performance of each of these representations (without fine-tuning) as inputs into a logistic classifier to predict 1-year mortality, healthcare use, and new disease diagnosis.

RESULTS

Patient representations generated by sequence-based algorithms performed consistently better than bag-of-words methods in predicting clinical endpoints, with the highest performance for EHR-BERT across all tasks, although the absolute improvement was small. Representations generated using disease categories perform similarly to those using diagnostic codes as inputs, suggesting models can equally manage smaller or larger vocabularies for prediction of these outcomes.

DISCUSSION AND CONCLUSION

Patient representations produced by sequence-based NLP algorithms from sequences of disease codes demonstrate improved predictive content for patient outcomes compared with representations generated by co-occurrence-based algorithms. This suggests transformer models may be useful for generating multi-purpose representations, even without fine-tuning.

A Qualitative Study of Physicians' Views on the Reuse of Electronic Health Record Data for Secondary Analysis

Electronic health records (EHRs) have become ubiquitous in clinical practice. Given the rich biomedical data captured for a large panel of patients, secondary analysis of these data for health research is also commonplace. Yet, there are many caveats to EHR data that the researchers must be aware of, such as the accuracy of and motive for documentation, and the reason for patients' visits to the clinic. The clinician-the author of the documentation-is thus central to the correct interpretation of EHR data for research purposes. In this study, I interviewed 11 physicians in various clinical specialties to bring attention to their view on the validity of research using EHR data. Qualitative, in-depth, one-on-one interviews were conducted with practicing physicians in inpatient and outpatient medicine. Content analysis using a data-driven, inductive approach to identify themes related to challenges and opportunities in the reuse of EHR data for secondary analysis generated seven themes. Themes that reflected challenges of EHRs for research included (1) audience, (2) accuracy of data, (3) availability of data, (4) documentation practices, and (5) representativeness. Themes that reflected opportunities of EHRs for research included (6) endorsement and (7) enablers. The greatest perceived barriers reflected the intended audience of the EHR, the interpretation and meaning of the data, and the quality of the data for research purposes. Physicians generally expressed more perceived challenges than opportunities in the reuse of EHR data for research purposes; however, they remained optimistic.

Enhancing data practices for Whole Health: Strategies for a transformative future

We explored the challenges and solutions for managing data within the Whole Health System (WHS), which operates as a Learning Health System and a patient-centered healthcare approach that combines conventional and complementary approaches. Addressing these challenges is critical for enhancing patient care and improving outcomes within WHS. The proposed solutions include prioritizing interoperability for seamless data exchange, incorporating patient-centered comparative clinical effectiveness research and real-world data to personalize treatment plans and validate integrative approaches, and leveraging advanced data analytics tools to incorporate patient-reported outcomes, objective metrics, robust data platforms. Implementing these measures will enable WHS to fulfill its mission as a holistic and patient-centered healthcare model, promoting greater collaboration among providers, boosting the well-being of patients and providers, and improving patient outcomes.

Quality of heart failure registration in primary care: observations from 1 million electronic health records in the Amsterdam Metropolitan Area

OBJECTIVES

Proper coding of heart failure (HF) in electronic health records (EHRs) is an important prerequisite for adequate care and research towards this vulnerable patient population. We set out to evaluate the accuracy of registration of HF diagnoses in primary care EHRs.

METHODS

In a routine primary care database covering the Amsterdam Metropolitan Area, we identified all episodes of care with International Classification of Primary Care (ICPC) codes K77 (decompensatio cordis) or K84.03 (cardiomyopathy) up to 31/12/2021. We also performed two text-based searches to identify HF episodes without an appropriate ICPC-code. An expert panel evaluated all ICPC and text matches for congruence between the assigned codes and notes.

RESULTS

From a database of 968,433 records we identified 19,106 patients (2.0 %) with a total of 24,011 ICPC-coded HF episodes. Removal of 1,324 episodes found to concern other or uncertain diagnoses and inclusion of 4,582 validated HF episodes identified through text search led to exclusion of 909 (overregistration: 4.8 %) and inclusion of 2,266 additional patients (underregistration: 11.1 %). The inclusion of miscoded HF episodes advanced the first known date of HF diagnosis in 3.9 % of records, with a median shift of 3.45 years. Episode-level underregistration decreased significantly over time, from 23.8 % in 2006 to 10.0 % in 2021.

CONCLUSIONS

While there is improvement over time, there are still substantial levels of over- and underregistration of HF, emphasizing the need for cautious interpretation of ICPC-coded data. The findings contribute to the understanding of HF registration issues in primary care and provide insights for improving registration practices.

Generalizability of a Musculoskeletal Therapist Electronic Health Record for Modelling Outcomes to Work-Related Musculoskeletal Disorders

PURPOSE

Electronic Health Records (EHRs) can contain vast amounts of clinical information that could be reused in modelling outcomes of work-related musculoskeletal disorders (WMSDs). Determining the generalizability of an EHR dataset is an important step in determining the appropriateness of its reuse. The study aims to describe the EHR dataset used by occupational musculoskeletal therapists and determine whether the EHR dataset is generalizable to the Australian workers' population and injury characteristics seen in workers' compensation claims.

METHODS

Variables were considered if they were associated with outcomes of WMSDs and variables data were available. Completeness and external validity assessment analysed frequency distributions, percentage of records and confidence intervals.

RESULTS

There were 48,434 patient care plans across 10 industries from 2014 to 2021. The EHR collects information related to clinical interventions, health and psychosocial factors, job demands, work accommodations as well as workplace culture, which have all been shown to be valuable variables in determining outcomes to WMSDs. Distributions of age, duration of employment, gender and region of birth were mostly similar to the Australian workforce. Upper limb WMSDs were higher in the EHR compared to workers' compensation claims and diagnoses were similar.

CONCLUSION

The study shows the EHR has strong potential to be used for further research into WMSDs as it has a similar population to the Australian workforce, manufacturing industry and workers' compensation claims. It contains many variables that may be relevant in modelling outcomes to WMSDs that are not typically available in existing datasets.

Overview and Clinical Applications of Artificial Intelligence and Machine Learning in Cardiac Anesthesiology

Artificial intelligence- (AI) and machine learning (ML)-based applications are becoming increasingly pervasive in the healthcare setting. This has in turn challenged clinicians, hospital administrators, and health policymakers to understand such technologies and develop frameworks for safe and sustained clinical implementation. Within cardiac anesthesiology, challenges and opportunities for AI/ML to support patient care are presented by the vast amounts of electronic health data, which are collected rapidly, interpreted, and acted upon within the periprocedural area. To address such challenges and opportunities, in this article, the authors review 3 recent applications relevant to cardiac anesthesiology, including depth of anesthesia monitoring, operating room resource optimization, and transthoracic/transesophageal echocardiography, as conceptual examples to explore strengths and limitations of AI/ML within healthcare, and characterize this evolving landscape. Through reviewing such applications, the authors introduce basic AI/ML concepts and methodologies, as well as practical considerations and ethical concerns for initiating and maintaining safe clinical implementation of AI/ML-based algorithms for cardiac anesthesia patient care.

Lessons learned from the development of a national registry on dementia care and support based on linked national health and administrative data

Introduction

This paper provides insight into the development of the Dutch Dementia Care and Support Registry and the lessons that can be learned from it. The aim of this Registry was to contribute to quality improvement in dementia care and support.

Methods

This paper describes how the Registry was set up in four stages, reflecting the four FAIR principles: the selection of data sources (Findability); obtaining access to the selected data sources (Accessibility); data linkage (Interoperability); and the reuse of data (Reusability).

Results

The linkage of 16 different data sources, including national routine health and administrative data appeared to be technically and legally feasible. The linked data in the Registry offers rich information about (the use of) care for persons with dementia across various healthcare settings, including but not limited to primary care, secondary care, long-term care and medication use, that cannot be obtained from single data sources.

Conclusions

A key lesson learned is that in order to reuse the data for quality improvement in practice, it is essential to involve healthcare professionals in setting up the Registry and to guide them in the interpretation of the data.

Incidence and management of mallet finger in Dutch primary care: a cohort study

BACKGROUND

A mallet finger (MF) is diagnosed clinically and can be managed in primary care. The actual incidence of MF and how it is managed in primary care is unknown.

AIM

To determine the incidence of MF in primary care and to obtain estimates for the proportions of osseous and tendon MF. An additional aim was to gain insight into the management of patients diagnosed with MF in primary care.

DESIGN & SETTING

A cohort study using a healthcare registration database from general practice in the Netherlands.

METHOD

Patients aged ≥18 years with a new diagnosis of MF from 1 January 2015-31 December 2019 were selected using a search algorithm based on International Classification of Primary Care (ICPC) coding.

RESULTS

In total, 161 cases of MF were identified. The mean incidence was 0.58 per 1000 person-years. A radiograph was taken in 58% (n = 93) of cases; 23% (n = 37) of cases had an osseous MF. The most applied strategies were referral to secondary care (45%) or conservative treatment in GP practice (43%). Overall, 7% were referred to a paramedical professional.

CONCLUSION

On average, a Dutch GP assesses ≥1 patient with MF per year. Since only a minimal number of patients required surgical treatment and a limited number of GPs requested radiography, the recommendation in the guidelines to perform radiography in all patients with MF should potentially be reconsidered. The purpose of requesting radiographs should not be to distinguish between a tendinogenic or osseous MF, but to assess whether there is a possible indication for surgery.

Challenges in and Opportunities for Electronic Health Record-Based Data Analysis and Interpretation

Electronic health records (EHRs) have been increasingly adopted in clinical practices across the United States, providing a primary source of data for clinical research, particularly observational cohort studies. EHRs are a high-yield, low-maintenance source of longitudinal real-world data for large patient populations and provide a wealth of information and clinical contexts that are useful for clinical research and translation into practice. Despite these strengths, it is important to recognize the multiple limitations and challenges related to the use of EHR data in clinical research. Missing data are a major source of error and biases and can affect the representativeness of the cohort of interest, as well as the accuracy of the outcomes and exposures. Here, we aim to provide a critical understanding of the types of data available in EHRs and describe the impact of data heterogeneity, quality, and generalizability, which should be evaluated prior to and during the analysis of EHR data. We also identify challenges pertaining to data quality, including errors and biases, and examine potential sources of such biases and errors. Finally, we discuss approaches to mitigate and remediate these limitations. A proactive approach to addressing these issues can help ensure the integrity and quality of EHR data and the appropriateness of their use in clinical studies.

Biases in Electronic Health Records Data for Generating Real-World Evidence: An Overview

Electronic Health Records (EHR) are increasingly being perceived as a unique source of data for clinical research as they provide unprecedentedly large volumes of real-time data from real-world settings. In this review of the secondary uses of EHR, we identify the anticipated breadth of opportunities, pointing out the data deficiencies and potential biases that are likely to limit the search for true causal relationships. This paper provides a comprehensive overview of the types of biases that arise along the pathways that generate real-world evidence and the sources of these biases. We distinguish between two levels in the production of EHR data where biases are likely to arise: (i) at the healthcare system level, where the principal source of bias resides in access to, and provision of, medical care, and in the acquisition and documentation of medical and administrative data; and (ii) at the research level, where biases arise from the processes of extracting, analyzing, and interpreting these data. Due to the plethora of biases, mainly in the form of selection and information bias, we conclude with advising extreme caution about making causal inferences based on secondary uses of EHRs.

COVID-19 waves in an urban setting 2020-2022: an electronic medical record analysis

Background

Global and national surveillance efforts have tracked COVID-19 incidence and clinical outcomes, but few studies have compared comorbid conditions and clinical outcomes across each wave of the pandemic. We analyzed data from the COVID-19 registry of a large urban healthcare system to determine the associations between presenting comorbidities and clinical outcomes during the pandemic.

Methods

We analyzed registry data for all inpatients and outpatients with COVID-19 from March 2020 through September 2022 (N = 44,499). Clinical outcomes were death, hospitalization, and intensive care unit (ICU) admission. Demographic and clinical outcomes data were analyzed overall and for each wave. Unadjusted and multivariable logistic regressions were performed to explore the associations between age, sex, race, ethnicity, comorbidities, and mortality.

Results

Waves 2 and 3 (Alpha and Delta variants) were associated with greater hospitalizations, ICU admissions, and mortality than other variants. Chronic pulmonary disease was the most common comorbid condition across all age groups and waves. Mortality rates were higher in older patients but decreased across all age groups in later waves. In every wave, mortality was associated with renal disease, congestive heart failure, cerebrovascular disease, diabetes, and chronic pulmonary disease. Multivariable analysis found that liver disease and renal disease were significantly associated with mortality, hospitalization, and ICU admission, and diabetes was significantly associated with hospitalization and ICU admission.

Conclusion

The COVID-19 registry is a valuable resource to identify risk factors for clinical outcomes. Our findings may inform risk stratification and care planning for patients with COVID-19 based on age and comorbid conditions.

"In conferences, everyone goes 'health data is the future' ": an interview study on challenges in re-using EHR data for research in Clinical Data Warehouses

More and more hospital Clinical Data Warehouses (CDWs) are developed to gain access to EHR data. The rapid growth of investments in CDWs suggest a real potential for innovation in healthcare. However, it is still not confirmed that CDWs will deliver on their promises as researchers working with CDWs face many challenges. To gain a better understanding of these challenges and how to overcome them, we conducted a series of semi-structured interviews with EHR data experts. In this article, we share some initial results from the ongoing interview study. Two main themes emerged from the analysis of the transcripts of the interviews: the importance of infrastructures in terms of data and how it is generated, and the difficulty to make care, clinical research, and data science work together. Finally, based on the experts' experience, several recommendations were identified when using a CDW.

Development and Validation of Case-Finding Algorithms for Digestive Cancer in the Spanish Healthcare Database BIFAP

BACKGROUND

electronic health records (EHRs) are helpful tools in epidemiology despite not being primarily collected for research. In Spain, primary care physicians play a central role and manage patients even in specialized care. All of this introduces variability that may lead to diagnostic inconsistencies. Therefore, data validation studies are crucial, so we aimed to develop and validate case-finding algorithms for digestive cancer in the primary care database BIFAP.

METHODS

from 2001 to 2019, subjects aged 40-89 without a cancer history were included. Case-finding algorithms using diagnostic codes and text-mining were built. We randomly sampled, clustered, and manually reviewed 816 EHRs. Then, positive predictive values (PPVs) and 95% confidence intervals (95% CIs) for each cancer were computed. Age and sex standardized incidence rates (SIRs) were compared with those reported by the National Cancer Registry (REDECAN).

RESULTS

we identified 95,672 potential cases. After validation, the PPV (95% CI) for hepato-biliary cancer was 87.6% (81.8-93.4), for esophageal cancer, it was 96.2% (93.1-99.2), for pancreatic cancer, it was 89.4% (84.5-94.3), for gastric cancer, it was 92.5% (88.3-96.6), and for colorectal cancer, it was 95.2% (92.1-98.4). The SIRs were comparable to those reported by the REDECAN.

CONCLUSIONS

the case-finding algorithms demonstrated high performance, supporting BIFAP as a suitable source of information to conduct epidemiologic studies of digestive cancer.

The Buzz Surrounding Precision Medicine: The Imperative of Incorporating It into Evidence-Based Medical Practice

Precision medicine (PM), through the integration of omics and environmental data, aims to provide a more precise prevention, diagnosis, and treatment of disease. Currently, PM is one of the emerging approaches in modern healthcare and public health, with wide implications for health care delivery, public health policy making formulation, and entrepreneurial endeavors. In spite of its growing popularity and the buzz surrounding it, PM is still in its nascent phase, facing considerable challenges that need to be addressed and resolved for it to attain the acclaim for which it strives. In this article, we discuss some of the current methodological pitfalls of PM, including the use of big data, and provide a perspective on how these challenges can be overcome by bringing PM closer to evidence-based medicine (EBM). Furthermore, to maximize the potential of PM, we present real-world illustrations of how EBM principles can be integrated into a PM approach.

A US real-world study of treatment patterns and outcomes in localized or locally advanced prostate cancer patients

PURPOSE

Men with localized or locally advanced prostate cancer (LPC/LAPC) are at risk of progression after radiotherapy (RT) or radical prostatectomy (RP). Using real-world data, we evaluated patient characteristics, treatment patterns, and outcomes in LPC/LAPC.

METHODS

Optum claims and electronic health records (EHR) data from January 2010 to December 2021 were queried for men with LPC/LAPC who received primary RT, RP, or androgen deprivation therapy alone within 180 days after diagnosis. Survival outcomes were analyzed using descriptive statistics and Kaplan-Meier curves. Real-world overall survival (rwOS) was compared in patients with and without evidence of disease (i.e., disease recurrence, metastasis, diagnosis of castration-resistant PC) at defined time points.

RESULTS

61,772 and 62,361 men in claims and EHR cohorts met the inclusion criteria. Median follow-up was 719 and 901 days, respectively. Most men received primary RT (51.0% claims, 35.0% EHR) or RP (39.4% claims, 53.8% EHR). Survival was greatest among men treated with RP, followed by RT. Adjusted for age and comorbidity, rwOS was shorter among men with evidence of disease within 1, 3, 4, and 5 years after primary treatment than those without at the same time points.

CONCLUSION

Real-world claims and EHR data show that survival among men with LPC/LAPC differs by primary treatment and time point of disease recurrence thereafter. Poor outcomes in men with LPC/LAPC who progress early indicate an unmet medical need for more effective primary treatment. If validated for surrogacy, no evidence of disease at specific time points could represent an intermediate efficacy endpoint in future trials.

Patellofemoral pain in general practice: the incidence and management

BACKGROUND

Patellofemoral pain (PFP) is a nontraumatic knee problem primarily observed in physically active adolescents. The objective of this study was to determine the incidence and management of PFP in children and adolescents in general practice.

METHODS

A retrospective cohort study was conducted using a regional primary care database containing full electronic health records of over 300,000 patients. Patients with a new PFP diagnosis between the years 2013 and 2019 were extracted using a search algorithm based on International Classification of Primary Health Care coding and search terms in free text. Data on the management of PFP were manually checked and analysed. In addition, a sub-analysis for chronic and nonchronic PFP patients was performed.

RESULTS

The mean incidence of PFP over the study period was 3.4 (95% CI 3.2-3.6) per 1,000 person years in the age group of 7-24 years. Girls had a higher incidence rate (4.6 [95% CI 4.3-5.0]) compared to boys (2.3 [95% CI 2.1-2.5]). Peak incidence was at age 13 years for both sexes. The most commonly applied management strategy was advice (55.1%), followed by referral to physiotherapy (28.2%), analgesics prescription (10.4%), and referral to the orthopaedic surgeon (8.9%). No differences were found in age, sex, and treatment between chronic and nonchronic PFP patients.

CONCLUSIONS

The average Dutch general practitioner sees approximately 1.4 new child or adolescent with PFP per year. Overall management strategies were in concordance with current Dutch general practice guideline on nontraumatic knee problems. More insight should be gained in the population with chronic complaints.

The Roles of Electronic Health Records for Clinical Trials in Low- and Middle-Income Countries: Scoping Review

BACKGROUND

Clinical trials are a crucial element in advancing medical knowledge and developing new treatments by establishing the evidence base for safety and therapeutic efficacy. However, the success of these trials depends on various factors, including trial design, project planning, research staff training, and adequate sample size. It is also crucial to recruit participants efficiently and retain them throughout the trial to ensure timely completion.

OBJECTIVE

There is an increasing interest in using electronic health records (EHRs)-a widely adopted tool in clinical practice-for clinical trials. This scoping review aims to understand the use of EHR in supporting the conduct of clinical trials in low- and middle-income countries (LMICs) and to identify its strengths and limitations.

METHODS

A comprehensive search was performed using 5 databases: MEDLINE, Embase, Scopus, Cochrane Library, and the Cumulative Index to Nursing and Allied Health Literature. We followed the latest version of the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guideline to conduct this review. We included clinical trials that used EHR at any step, conducted a narrative synthesis of the included studies, and mapped the roles of EHRs into the life cycle of a clinical trial.

RESULTS

A total of 30 studies met the inclusion criteria: 13 were randomized controlled trials, 3 were cluster randomized controlled trials, 12 were quasi-experimental studies, and 2 were feasibility pilot studies. Most of the studies addressed infectious diseases (15/30, 50%), with 80% (12/15) of them about HIV or AIDS and another 40% (12/30) focused on noncommunicable diseases. Our synthesis divided the roles of EHRs into 7 major categories: participant identification and recruitment (12/30, 40%), baseline information collection (6/30, 20%), intervention (8/30, 27%), fidelity assessment (2/30, 7%), primary outcome assessment (24/30, 80%), nonprimary outcome assessment (13/30, 43%), and extended follow-up (2/30, 7%). None of the studies used EHR for participant consent and randomization.

CONCLUSIONS

Despite the enormous potential of EHRs to increase the effectiveness and efficiency of conducting clinical trials in LMICs, challenges remain. Continued exploration of the appropriate uses of EHRs by navigating their strengths and limitations to ensure fitness for use is necessary to better understand the most optimal uses of EHRs for conducting clinical trials in LMICs.

Database quality assessment in research in paramedicine: a scoping review

BACKGROUND

Research in paramedicine faces challenges in developing research capacity, including access to high-quality data. A variety of unique factors in the paramedic work environment influence data quality. In other fields of healthcare, data quality assessment (DQA) frameworks provide common methods of quality assessment as well as standards of transparent reporting. No similar DQA frameworks exist for paramedicine, and practices related to DQA are sporadically reported. This scoping review aims to describe the range, extent, and nature of DQA practices within research in paramedicine.

METHODS

This review followed a registered and published protocol. In consultation with a professional librarian, a search strategy was developed and applied to MEDLINE (National Library of Medicine), EMBASE (Elsevier), Scopus (Elsevier), and CINAHL (EBSCO) to identify studies published from 2011 through 2021 that assess paramedic data quality as a stated goal. Studies that reported quantitative results of DQA using data that relate primarily to the paramedic practice environment were included. Protocols, commentaries, and similar study types were excluded. Title/abstract screening was conducted by two reviewers; full-text screening was conducted by two, with a third participating to resolve disagreements. Data were extracted using a piloted data-charting form.

RESULTS

Searching yielded 10,105 unique articles. After title and abstract screening, 199 remained for full-text review; 97 were included in the analysis. Included studies varied widely in many characteristics. Majorities were conducted in the United States (51%), assessed data containing between 100 and 9,999 records (61%), or assessed one of three topic areas: data, trauma, or out-of-hospital cardiac arrest (61%). All data-quality domains assessed could be grouped under 5 summary domains: completeness, linkage, accuracy, reliability, and representativeness.

CONCLUSIONS

There are few common standards in terms of variables, domains, methods, or quality thresholds for DQA in paramedic research. Terminology used to describe quality domains varied among included studies and frequently overlapped. The included studies showed no evidence of assessing some domains and emerging topics seen in other areas of healthcare. Research in paramedicine would benefit from a standardized framework for DQA that allows for local variation while establishing common methods, terminology, and reporting standards.

Severity of Prior Coronavirus Disease 2019 is Associated With Postoperative Outcomes After Major Inpatient Surgery

OBJECTIVE

To determine how the severity of prior history (Hx) of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) infection influences postoperative outcomes after major elective inpatient surgery.

BACKGROUND

Surgical guidelines instituted early in the coronavirus disease 2019 (COVID-19) pandemic recommended a delay in surgery of up to 8 weeks after an acute SARS-CoV-2 infection. This was based on the observation of elevated surgical risk after recovery from COVID-19 early in the pandemic. As the pandemic shifts to an endemic phase, it is unclear whether this association remains, especially for those recovering from asymptomatic or mildly symptomatic COVID-19.

METHODS

Utilizing the National COVID Cohort Collaborative, we assessed postoperative outcomes for adults with and without a Hx of COVID-19 who underwent major elective inpatient surgery between January 2020 and February 2023. COVID-19 severity and time from infection to surgery were each used as independent variables in multivariable logistic regression models.

RESULTS

This study included 387,030 patients, of whom 37,354 (9.7%) were diagnosed with preoperative COVID-19. Hx of COVID-19 was found to be an independent risk factor for adverse postoperative outcomes even after a 12-week delay for patients with moderate and severe SARS-CoV-2 infection. Patients with mild COVID-19 did not have an increased risk of adverse postoperative outcomes at any time point. Vaccination decreased the odds of respiratory failure.

CONCLUSIONS

Impact of COVID-19 on postoperative outcomes is dependent on the severity of illness, with only moderate and severe disease leading to a higher risk of adverse outcomes. Existing perioperative policies should be updated to include consideration of COVID-19 disease severity and vaccination status.

Second-Line Pharmaceutical Treatments for Patients with Type 2 Diabetes

Importance

Assessing the relative effectiveness and safety of additional treatments when metformin monotherapy is insufficient remains a limiting factor in improving treatment choices in type 2 diabetes.

Objective

To determine whether data from electronic health records across the University of California Health system could be used to assess the comparative effectiveness and safety associated with 4 treatments in diabetes when added to metformin monotherapy.

Design, Setting, and Participants

This multicenter, new user, multidimensional propensity score-matched retrospective cohort study with leave-one-medical-center-out (LOMCO) sensitivity analysis used principles of emulating target trial. Participants included patients with diabetes receiving metformin who were then additionally prescribed either a sulfonylurea, dipeptidyl peptidase-4 inhibitor (DPP4I), sodium-glucose cotransporter-2 inhibitor (SGLT2I), or glucagon-like peptide-1 receptor agonist (GLP1RA) for the first time and followed-up over a 5-year monitoring period. Data were analyzed between January 2022 and April 2023.

Exposure

Treatment with sulfonylurea, DPP4I, SGLT2I, or GLP1RA added to metformin monotherapy.

Main Outcomes and Measures

The main effectiveness outcome was the ability of patients to maintain glycemic control, represented as time to metabolic failure (hemoglobin A1c [HbA1c] ≥7.0%). A secondary effectiveness outcome was assessed by monitoring time to new incidence of any of 28 adverse outcomes, including diabetes-related complications while treated with the assigned drug. Sensitivity analysis included LOMCO.

Results

This cohort study included 31 852 patients (16 635 [52.2%] male; mean [SD] age, 61.4 [12.6] years) who were new users of diabetes treatments added on to metformin monotherapy. Compared with sulfonylurea in random-effect meta-analysis, treatment with SGLT2I (summary hazard ratio [sHR], 0.75 [95% CI, 0.69-0.83]; I2 = 37.5%), DPP4I (sHR, 0.79 [95% CI, 0.75-0.84]; I2 = 0%), GLP1RA (sHR, 0.62 [95% CI, 0.57-0.68]; I2 = 23.6%) were effective in glycemic control; findings from LOMCO sensitivity analysis were similar. Treatment with SGLT2I showed no significant difference in effectiveness compared with GLP1RA (sHR, 1.26 [95% CI, 1.12-1.42]; I2 = 47.3%; no LOMCO) or DPP4I (sHR, 0.97 [95% CI, 0.90-1.04]; I2 = 0%). Patients treated with DPP4I and SGLT2I had fewer cardiovascular events compared with those treated with sulfonylurea (DPP4I: sHR, 0.84 [95% CI, 0.74-0.96]; I2 = 0%; SGLT2I: sHR, 0.78 [95% CI, 0.62-0.98]; I2 = 0%). Patients treated with a GLP1RA or SGLT2I were less likely to develop chronic kidney disease (GLP1RA: sHR, 0.75 [95% CI 0.6-0.94]; I2 = 0%; SGLT2I: sHR, 0.77 [95% CI, 0.61-0.97]; I2 = 0%), kidney failure (GLP1RA: sHR, 0.69 [95% CI, 0.56-0.86]; I2 = 9.1%; SGLT2I: sHR, 0.72 [95% CI, 0.59-0.88]; I2 = 0%), or hypertension (GLP1RA: sHR, 0.82 [95% CI, 0.68-0.97]; I2 = 0%; SGLT2I: sHR, 0.73 [95% CI, 0.58-0.92]; I2 = 38.5%) compared with those treated with a sulfonylurea. Patients treated with an SGLT2I, vs a DPP4I, GLP1RA, or sulfonylurea, were less likely to develop indicators of chronic hepatic dysfunction (sHR vs DPP4I, 0.68 [95% CI, 0.49-0.95]; I2 = 0%; sHR vs GLP1RA, 0.66 [95% CI, 0.48-0.91]; I2 = 0%; sHR vs sulfonylurea, 0.60 [95% CI, 0.44-0.81]; I2 = 0%), and those treated with a DPP4I were less likely to develop new incidence of hypoglycemia (sHR, 0.48 [95% CI, 0.36-0.65]; I2 = 22.7%) compared with those treated with a sulfonylurea.

Conclusions and Relevance

These findings highlight familiar medication patterns, including those mirroring randomized clinical trials, as well as providing new insights underscoring the value of robust clinical data analytics in swiftly generating evidence to help guide treatment choices in diabetes.

Mapping and evaluating national data flows: transparency, privacy, and guiding infrastructural transformation

The importance of big health data is recognised worldwide. Most UK National Health Service (NHS) care interactions are recorded in electronic health records, resulting in an unmatched potential for population-level datasets. However, policy reviews have highlighted challenges from a complex data-sharing landscape relating to transparency, privacy, and analysis capabilities. In response, we used public information sources to map all electronic patient data flows across England, from providers to more than 460 subsequent academic, commercial, and public data consumers. Although NHS data support a global research ecosystem, we found that multistage data flow chains limit transparency and risk public trust, most data interactions do not fulfil recommended best practices for safe data access, and existing infrastructure produces aggregation of duplicate data assets, thus limiting diversity of data and added value to end users. We provide recommendations to support data infrastructure transformation and have produced a website (https://DataInsights.uk) to promote transparency and showcase NHS data assets.

Identifying potential biases in code sequences in primary care electronic healthcare records: a retrospective cohort study of the determinants of code frequency

OBJECTIVES

To determine whether the frequency of diagnostic codes for long-term conditions (LTCs) in primary care electronic healthcare records (EHRs) is associated with (1) disease coding incentives, (2) General Practice (GP), (3) patient sociodemographic characteristics and (4) calendar year of diagnosis.

DESIGN

Retrospective cohort study.

SETTING

GPs in England from 2015 to 2022 contributing to the Clinical Practice Research Datalink Aurum dataset.

PARTICIPANTS

All patients registered to a GP with at least one incident LTC diagnosed between 1 January 2015 and 31 December 2019.

PRIMARY AND SECONDARY OUTCOME MEASURES

The number of diagnostic codes for an LTC in (1) the first and (2) the second year following diagnosis, stratified by inclusion in the Quality and Outcomes Framework (QOF) financial incentive programme.

RESULTS

3 113 724 patients were included, with 7 723 365 incident LTCs. Conditions included in QOF had higher rates of annual coding than conditions not included in QOF (1.03 vs 0.32 per year, p<0.0001). There was significant variation in code frequency by GP which was not explained by patient sociodemographics. We found significant associations with patient sociodemographics, with a trend towards higher coding rates in people living in areas of higher deprivation for both QOF and non-QOF conditions. Code frequency was lower for conditions with follow-up time in 2020, associated with the onset of the COVID-19 pandemic.

CONCLUSIONS

The frequency of diagnostic codes for newly diagnosed LTCs is influenced by factors including patient sociodemographics, disease inclusion in QOF, GP practice and the impact of the COVID-19 pandemic. Natural language processing or other methods using temporally ordered code sequences should account for these factors to minimise potential bias.

Self-Harm in Children and Adolescents Who Presented at Emergency Units During the COVID-19 Pandemic: An International Retrospective Cohort Study

OBJECTIVE

To compare psychiatric emergencies and self-harm at emergency departments (EDs) 1 year into the pandemic, to early pandemic and pre-pandemic, and to examine the changes in the characteristics of self-harm presentations.

METHOD

This retrospective cohort study expanded on the Pandemic-Related Emergency Psychiatric Presentations (PREP-kids) study. Routine record data in March to April of 2019, 2020, and 2021 from 62 EDs in 25 countries were included. ED presentations made by children and adolescents for any mental health reasons were analyzed.

RESULTS

Altogether, 8,174 psychiatric presentations were recorded (63.5% female; mean [SD] age, 14.3 [2.6] years), 3,742 of which were self-harm presentations. Rate of psychiatric ED presentations in March to April 2021 was twice as high as in March to April 2020 (incidence rate ratio [IRR], 1.93; 95% CI, 1.60-2.33), and 50% higher than in March to April 2019 (IRR, 1.51; 95% CI, 1.25-1.81). Rate of self-harm presentations doubled between March to April 2020 and March to April 2021 (IRR, 1.98; 95% CI, 1.68-2.34), and was overall 1.7 times higher than in March to April 2019 (IRR, 1.70; 95% CI, 1.44-2.00). Comparing self-harm characteristics in March to April 2021 with March to April 2019, self-harm contributed to a higher proportion of all psychiatric presentations (odds ratio [OR], 1.30; 95% CI, 1.05-1.62), whereas female representation in self-harm presentations doubled (OR, 1.98; 95% CI, 1.45-2.72) and follow-up appointments were offered 4 times as often (OR, 4.46; 95% CI, 2.32-8.58).

CONCLUSION

Increased pediatric ED visits for both self-harm and psychiatric reasons were observed, suggesting potential deterioration in child mental health. Self-harm in girls possibly increased and needs to be prioritized. Clinical services should continue using follow-up appointments to support discharge from EDs.

DIVERSITY & INCLUSION STATEMENT

One or more of the authors of this paper self-identifies as a member of one or more historically underrepresented racial and/or ethnic groups in science. We actively worked to promote inclusion of historically underrepresented racial and/or ethnic groups in science in our author group. While citing references scientifically relevant for this work, we also actively worked to promote inclusion of historically underrepresented racial and/or ethnic groups in science in our reference list. The author list of this paper includes contributors from the location and/or community where the research was conducted who participated in the data collection, design, analysis, and/or interpretation of the work.

Diagnostic rate estimation from Medicare records: Dependence on claim numbers and latent clinical features

OBJECTIVE

International Classification of Disorders version 10 (ICD-10) codes contribute heavily to healthcare data. Medicare claims and other data-sources are used to constitute study populations and appraise healthcare processes. How variability in claims-per-beneficiary impacts diagnostic determinations is inadequately understood. The objective of this study is so assess distributional properties of Medicare claims, and examine claim rates impact on code utilization and rate determinations.

METHODS

The study population was Medicare beneficiaries aged 75-79.99 with claim(s) in the 5% standard analytical Carrier and Outpatient files, alive and participating in Medicare part B for all 12 months of 2017. Medicare beneficiary files were processed to create records containing all ICD-10 codes specified, key demographics, Part B and vital status, and the total claims for each 2017 beneficiary. Claim number cohorts were characterized.

RESULTS

Beneficiaries meeting inclusion criteria totaled 221,625, these having 7,617,503 claims; 96.4% had between 1 and 120 claims. Median claims were 24 for males (females 25); modal claims were 11 (13). Average distinct codes per beneficiary increased with claims number. The assignment of ICD-10 codes, i.e., 'diagnostic rate estimates' (DRE), increased as claim numbers increased for most codes among those most commonly utilized. For some conditions, mostly benign and age-related, DREs plateaued as claim numbers increased. For other conditions, typically associated with clinical acuity, e.g., chest pain, DREs increased steeply with claims.

CONCLUSIONS

Older adult Medicare beneficiaries aged 75-80 exhibited varying claims activity over the course of a year. Although DRE dependence on claim numbers varies across ICD-10 codes, rate estimates are higher for beneficiaries with claim numbers above the median.

lab: an R package for generating analysis-ready data from laboratory records

Background

Electronic health records (EHRs) play a crucial role in healthcare decision-making by giving physicians insights into disease progression and suitable treatment options. Within EHRs, laboratory test results are frequently utilized for predicting disease progression. However, processing laboratory test results often poses challenges due to variations in units and formats. In addition, leveraging the temporal information in EHRs can improve outcomes, prognoses, and diagnosis predication. Nevertheless, the irregular frequency of the data in these records necessitates data preprocessing, which can add complexity to time-series analyses.

Methods

To address these challenges, we developed an open-source R package that facilitates the extraction of temporal information from laboratory records. The proposed lab package generates analysis-ready time series data by segmenting the data into time-series windows and imputing missing values. Moreover, users can map local laboratory codes to the Logical Observation Identifier Names and Codes (LOINC), an international standard. This mapping allows users to incorporate additional information, such as reference ranges and related diseases. Moreover, the reference ranges provided by LOINC enable us to categorize results into normal or abnormal. Finally, the analysis-ready time series data can be further summarized using descriptive statistics and utilized to develop models using machine learning technologies.

Results

Using the lab package, we analyzed data from MIMIC-III, focusing on newborns with patent ductus arteriosus (PDA). We extracted time-series laboratory records and compared the differences in test results between patients with and without 30-day in-hospital mortality. We then identified significant variations in several laboratory test results 7 days after PDA diagnosis. Leveraging the time series-analysis-ready data, we trained a prediction model with the long short-term memory algorithm, achieving an area under the receiver operating characteristic curve of 0.83 for predicting 30-day in-hospital mortality in model training. These findings demonstrate the lab package's effectiveness in analyzing disease progression.

Conclusions

The proposed lab package simplifies and expedites the workflow involved in laboratory records extraction. This tool is particularly valuable in assisting clinical data analysts in overcoming the obstacles associated with heterogeneous and sparse laboratory records.

The Constrained-Disorder Principle Assists in Overcoming Significant Challenges in Digital Health: Moving from "Nice to Have" to Mandatory Systems

The success of artificial intelligence depends on whether it can penetrate the boundaries of evidence-based medicine, the lack of policies, and the resistance of medical professionals to its use. The failure of digital health to meet expectations requires rethinking some of the challenges faced. We discuss some of the most significant challenges faced by patients, physicians, payers, pharmaceutical companies, and health systems in the digital world. The goal of healthcare systems is to improve outcomes. Assisting in diagnosing, collecting data, and simplifying processes is a "nice to have" tool, but it is not essential. Many of these systems have yet to be shown to improve outcomes. Current outcome-based expectations and economic constraints make "nice to have," "assists," and "ease processes" insufficient. Complex biological systems are defined by their inherent disorder, bounded by dynamic boundaries, as described by the constrained disorder principle (CDP). It provides a platform for correcting systems' malfunctions by regulating their degree of variability. A CDP-based second-generation artificial intelligence system provides solutions to some challenges digital health faces. Therapeutic interventions are held to improve outcomes with these systems. In addition to improving clinically meaningful endpoints, CDP-based second-generation algorithms ensure patient and physician engagement and reduce the health system's costs.

Potential bias and lack of generalizability in electronic health record data: reflections on health equity from the National Institutes of Health Pragmatic Trials Collaboratory

Embedded pragmatic clinical trials (ePCTs) play a vital role in addressing current population health problems, and their use of electronic health record (EHR) systems promises efficiencies that will increase the speed and volume of relevant and generalizable research. However, as the number of ePCTs using EHR-derived data grows, so does the risk that research will become more vulnerable to biases due to differences in data capture and access to care for different subsets of the population, thereby propagating inequities in health and the healthcare system. We identify 3 challenges-incomplete and variable capture of data on social determinants of health, lack of representation of vulnerable populations that do not access or receive treatment, and data loss due to variable use of technology-that exacerbate bias when working with EHR data and offer recommendations and examples of ways to actively mitigate bias.

SCOPE: predicting future diagnoses in office visits using electronic health records

We propose an interpretable and scalable model to predict likely diagnoses at an encounter based on past diagnoses and lab results. This model is intended to aid physicians in their interaction with the electronic health records (EHR). To accomplish this, we retrospectively collected and de-identified EHR data of 2,701,522 patients at Stanford Healthcare over a time period from January 2008 to December 2016. A population-based sample of patients comprising 524,198 individuals (44% M, 56% F) with multiple encounters with at least one frequently occurring diagnosis codes were chosen. A calibrated model was developed to predict ICD-10 diagnosis codes at an encounter based on the past diagnoses and lab results, using a binary relevance based multi-label modeling strategy. Logistic regression and random forests were tested as the base classifier, and several time windows were tested for aggregating the past diagnoses and labs. This modeling approach was compared to a recurrent neural network based deep learning method. The best model used random forest as the base classifier and integrated demographic features, diagnosis codes, and lab results. The best model was calibrated and its performance was comparable or better than existing methods in terms of various metrics, including a median AUROC of 0.904 (IQR [0.838, 0.954]) over 583 diseases. When predicting the first occurrence of a disease label for a patient, the median AUROC with the best model was 0.796 (IQR [0.737, 0.868]). Our modeling approach performed comparably as the tested deep learning method, outperforming it in terms of AUROC (p < 0.001) but underperforming in terms of AUPRC (p < 0.001). Interpreting the model showed that the model uses meaningful features and highlights many interesting associations among diagnoses and lab results. We conclude that the multi-label model performs comparably with RNN based deep learning model while offering simplicity and potentially superior interpretability. While the model was trained and validated on data obtained from a single institution, its simplicity, interpretability and performance makes it a promising candidate for deployment.

Procurement of patient medical records from multiple health care facilities for public health research: feasibility, challenges, and lessons learned

Objectives

Studies that combine medical record and primary data are typically conducted in a small number of health care facilities (HCFs) covering a limited catchment area; however, depending on the study objectives, validity may be improved by recruiting a more expansive sample of patients receiving care across multiple HCFs. We evaluate the feasibility of a novel protocol to obtain patient medical records from multiple HCFs using a broad representative sampling frame.

Materials and Methods

In a prospective cohort study on HIV pre-exposure prophylaxis utilization, primary data were collected from a representative sample of community-dwelling participants; voluntary authorization was obtained to access participants' medical records from the HCF at which they were receiving care. Medical record procurement procedures were documented for later analysis.

Results

The cohort consisted of 460 participants receiving care from 122 HCFs; 81 participants were lost to follow-up resulting in 379 requests for medical records submitted to HCFs, and a total of 343 medical records were obtained (91% response rate). Less than 20% of the medical records received were in electronic form. On average, the cost of medical record acquisition was $120 USD per medical record.

Conclusions

Obtaining medical record data on research participants receiving care across multiple HCFs was feasible, but time-consuming and resulted in appreciable missing data. Researchers combining primary data with medical record data should select a sampling and data collection approach that optimizes study validity while weighing the potential benefits (more representative sample; inclusion of HCF-level predictors) and drawbacks (cost, missing data) of obtaining medical records from multiple HCFs.

Healthcare utilization patterns for knee or hip osteoarthritis before and after changes in national health insurance coverage: A data linkage study

INTRODUCTION

Medical guidelines aim to stimulate stepped care for knee and hip osteoarthritis, redirecting treatments from hospitals to primary care. In the Netherlands, this development was supported by changing health insurance coverage for physio/exercise therapy. The aim of this study was to evaluate healthcare utilization patterns before and after health changes in health insurance coverage.

METHOD

We analyzed electronic health records and claims data from patients with osteoarthritis in the knee (N = 32,091) and hip (N = 16,313). Changes between 2013 and 2019 in the proportion of patients treated by the general practitioner, physio/exercise therapist or orthopedic surgeon within 6 months after onset were assessed.

RESULTS

Joint replacement surgeries decreased for knee (OR 0.47 [0.41-0.54]) and hip (OR 0.81 [0.71-0.93]) osteoarthritis between 2013-2019. The use of physio/exercise therapy increased (knee: OR 1.38 [1.24-1.53], hip: OR 1.26 [1.08-1.47]). However, the proportion treated by a physio/exercise therapist decreased for patients that had not depleted their annual deductibles (knee: OR 0.86 [0.79 - 0.94], hip: OR 0.90 [0.79 - 1.02]). This might be affected by the inclusion of physio/exercise therapy in basic health insurance in 2018.

CONCLUSION

We have found a shift from hospitals to primary care in knee and hip osteoarthritis care. However, the use of physio/exercise therapy declined after changes in insurance coverage for patients that had not depleted their deductibles.

Severity of Prior COVID-19 Infection is Associated with Postoperative Outcomes Following Major Inpatient Surgery

Objective

To determine the association between severity of prior history of SARS-CoV-2 infection and postoperative outcomes following major elective inpatient surgery.

Summary Background Data

Surgical guidelines instituted early in the COVID-19 pandemic recommended delay in surgery up to 8 weeks following an acute SARS-CoV-2 infection. Given that surgical delay can lead to worse medical outcomes, it is unclear if continuation of such stringent policies is necessary and beneficial for all patients, especially those recovering from asymptomatic or mildly symptomatic COVID-19.

Methods

Utilizing the National Covid Cohort Collaborative (N3C), we assessed postoperative outcomes for adults with and without a history of COVID-19 who underwent major elective inpatient surgery between January 2020 and February 2023. COVID-19 severity and time from SARS-CoV-2 infection to surgery were each used as independent variables in multivariable logistic regression models.

Results

This study included 387,030 patients, of which 37,354 (9.7%) had a diagnosis of preoperative COVID-19. History of COVID-19 was found to be an independent risk factor for adverse postoperative outcomes even after a 12-week delay for patients with moderate and severe SARS-CoV-2 infection. Patients with mild COVID-19 did not have an increased risk of adverse postoperative outcomes at any time point. Vaccination decreased the odds of mortality and other complications.

Conclusions

Impact of COVID-19 on postoperative outcomes is dependent on severity of illness, with only moderate and severe disease leading to higher risk of adverse outcomes. Existing wait time policies should be updated to include consideration of COVID-19 disease severity and vaccination status.

Care by general practitioners for patients with asthma or COPD during the COVID-19 pandemic

The impact of the COVID-19 pandemic on general practitioners' (GP) care for patients with asthma and/or COPD is largely unknown. To describe the impact of the pandemic on asthma or COPD-related GP care, we analysed routinely recorded electronic health records data from Dutch general practices and out-of-hours (OOH) services. During the COVID-19 pandemic (2020), the contact rates for asthma and/or COPD were significantly lower in GP practices and OOH services compared with the pre-pandemic period (2019) (respectively, 15% lower and 28% lower). The proportion of telephone contacts increased significantly with 13%-point in GP practices and 12%-point at OOH services, while the proportion of face-to-face contacts decreased. Furthermore, the proportion of high urgent contacts with OOH services decreased by 8.5%-point. To conclude, the overall contact rates in GP practices and OOH services decreased, while more contacts were remote. Lower contact rates have, after a short follow-up, not resulted in more patients with exacerbations in OOH care. However, this might still be expected after a longer follow-up.

Occurrence of comorbidity following osteoarthritis diagnosis: a cohort study in the Netherlands

OBJECTIVE

To determine the risk of comorbidity following diagnosis of knee or hip osteoarthritis (OA).

DESIGN

A cohort study was conducted using the Integrated Primary Care Information database, containing electronic health records of 2.5 million patients from the Netherlands. Adults at risk for OA were included. Diagnosis of knee or hip OA (=exposure) and 58 long-term comorbidities (=outcome) were defined by diagnostic codes following the International Classification of Primary Care coding system. Time between the start of follow-up and incident diagnosis of OA was defined as unexposed, and between diagnosis of OA and the end of follow-up as exposed. Age and sex adjusted hazard ratios (HRs) comparing comorbidity rates in exposed and unexposed patient time were estimated with 99.9% confidence intervals (CI).

RESULTS

The study population consisted of 1,890,712 patients. For 30 of the 58 studied comorbidities, exposure to knee OA showed a HR larger than 1. Largest positive associations (HR with (99.9% CIs)) were found for obesity 2.55 (2.29-2.84) and fibromyalgia 2.06 (1.53-2.77). For two conditions a HR < 1 was found, other comorbidities showed no association with exposure to knee OA. For 26 comorbidities, exposure to hip OA showed a HR larger than 1. The largest were found for polymyalgia rheumatica 1.81 (1.41-2.32) and fibromyalgia 1.70 (1.10-2.63). All other comorbidities showed no associations with hip OA.

CONCLUSION

This study showed that many comorbidities were diagnosed more often in patients with knee or hip OA. This suggests that the management of OA should consider the risk of other long-term-conditions.

Real-life and real-time hearing aid experiences: Insights from self-initiated ecological momentary assessments and natural language analysis

Introduction

Smartphone technology can provide an effective means to bring real-life and (near-)real-time feedback from hearing aid wearers into the clinic. Ecological Momentary Assessment (EMA) encourages listeners to report on their experiences during or shortly after they take place in order to minimize recall bias, e.g., guided by surveys in a mobile application. Allowing listeners to describe experiences in their own words, further, ensures that answers are independent of predefined jargon or of how survey questions are formulated. Through these means, one can obtain ecologically valid sets of data, for instance during a hearing aid trial, which can support clinicians to assess the needs of their clients, provide directions for fine-tuning, and counselling. At a larger scale, such datasets would facilitate training of machine learning algorithms that could help hearing technology to anticipate user needs.

Methods

In this retrospective, exploratory analysis of a clinical data set, we performed a cluster analysis on 8,793 open-text statements, which were collected through self-initiated EMAs, provided by 2,301 hearing aid wearers as part of their hearing care. Our aim was to explore how listeners describe their daily life experiences with hearing technology in (near-)real-time, in their own words, by identifying emerging themes in the reports. We also explored whether identified themes correlated with the nature of the experiences, i.e., self-reported satisfaction ratings indicating a positive or negative experience.

Results

Results showed that close to 60% of listeners' reports related to speech intelligibility in challenging situations and sound quality dimensions, and tended to be valued as positive experiences. In comparison, close to 40% of reports related to hearing aid management, and tended to be valued as negative experiences.

Discussion

This first report of open-text statements, collected through self-initiated EMAs as part of clinical practice, shows that, while EMA can come with a participant burden, at least a subsample of motivated hearing aid wearers could use these novel tools to provide feedback to inform more responsive, personalized, and family-centered hearing care.

The quality of vital signs measurements and value preferences in electronic medical records varies by hospital, specialty, and patient demographics

We aimed to assess the frequency of value preferences in recording of vital signs in electronic healthcare records (EHRs) and associated patient and hospital factors. We used EHR data from Oxford University Hospitals, UK, between 01-January-2016 and 30-June-2019 and a maximum likelihood estimator to determine the prevalence of value preferences in measurements of systolic and diastolic blood pressure (SBP/DBP), heart rate (HR) (readings ending in zero), respiratory rate (multiples of 2 or 4), and temperature (readings of 36.0 °C). We used multivariable logistic regression to investigate associations between value preferences and patient age, sex, ethnicity, deprivation, comorbidities, calendar time, hour of day, days into admission, hospital, day of week and speciality. In 4,375,654 records from 135,173 patients, there was an excess of temperature readings of 36.0 °C above that expected from the underlying distribution that affected 11.3% (95% CI 10.6-12.1%) of measurements, i.e. these observations were likely inappropriately recorded as 36.0 °C instead of the true value. SBP, DBP and HR were rounded to the nearest 10 in 2.2% (1.4-2.8%) and 2.0% (1.3-5.1%) and 2.4% (1.7-3.1%) of measurements. RR was also more commonly recorded as multiples of 2. BP digit preference and an excess of temperature recordings of 36.0 °C were more common in older and male patients, as length of stay increased, following a previous normal set of vital signs and typically more common in medical vs. surgical specialities. Differences were seen between hospitals, however, digit preference reduced over calendar time. Vital signs may not always be accurately documented, and this may vary by patient groups and hospital settings. Allowances and adjustments may be needed in delivering care to patients and in observational analyses and predictive tools using these factors as outcomes or exposures.

Opt-In and Opt-Out Consent Procedures for the Reuse of Routinely Recorded Health Data in Scientific Research and Their Consequences for Consent Rate and Consent Bias: Systematic Review

BACKGROUND

Scientific researchers who wish to reuse health data pertaining to individuals can obtain consent through an opt-in procedure or opt-out procedure. The choice of procedure may have consequences for the consent rate and representativeness of the study sample and the quality of the research, but these consequences are not well known.

OBJECTIVE

This review aimed to provide insight into the consequences for the consent rate and consent bias of the study sample of opt-in procedures versus opt-out procedures for the reuse of routinely recorded health data for scientific research purposes.

METHODS

A systematic review was performed based on searches in PubMed, Embase, CINAHL, PsycINFO, Web of Science Core Collection, and the Cochrane Library. Two reviewers independently included studies based on predefined eligibility criteria and assessed whether the statistical methods used in the reviewed literature were appropriate for describing the differences between consenters and nonconsenters. Statistical pooling was conducted, and a description of the results was provided.

RESULTS

A total of 15 studies were included in this meta-analysis. Of the 15 studies, 13 (87%) implemented an opt-in procedure, 1 (7%) implemented an opt-out procedure, and 1 (7%) implemented both the procedures. The average weighted consent rate was 84% (60,800/72,418 among the studies that used an opt-in procedure and 96.8% (2384/2463) in the single study that used an opt-out procedure. In the single study that described both procedures, the consent rate was 21% in the opt-in group and 95.6% in the opt-out group. Opt-in procedures resulted in more consent bias compared with opt-out procedures. In studies with an opt-in procedure, consenting individuals were more likely to be males, had a higher level of education, higher income, and higher socioeconomic status.

CONCLUSIONS

Consent rates are generally lower when using an opt-in procedure compared with using an opt-out procedure. Furthermore, in studies with an opt-in procedure, participants are less representative of the study population. However, both the study populations and the way in which opt-in or opt-out procedures were organized varied widely between the studies, which makes it difficult to draw general conclusions regarding the desired balance between patient control over data and learning from health data. The reuse of routinely recorded health data for scientific research purposes may be hampered by administrative burdens and the risk of bias.

Antibiotic Utilization during COVID-19: Are We Over-Prescribing?

The aims of this study were to analyze the utilization of antibiotics before (2018, 2019) and during the COVID-19 pandemic (2020) and the practice of prescribing antibiotics in outpatient settings for COVID-19 patients during the 2020-2022 period. The Anatomical Therapeutic Chemical Classification/Defined Daily Dose methodology was used for the analysis of outpatient antibiotic utilization in the Republic of Srpska. The data was expressed in DDD/1000 inhabitants/day. The rate of antibiotics prescribed to COVID-19 outpatients was analyzed using medical record data from 16,565 patients registered with B34.2, U07.1, and U07.2 World Health Organization International Classification of Diseases 10th revision codes. During 2020, outpatient antibiotic utilization increased by 53.80% compared to 2019. At least one antibiotic was prescribed for 91.04%, 83.05%, and 73.52% of COVID-19 outpatients during 2020, 2021, and the first half of 2022, respectively. On a monthly basis, at least one antibiotic was prescribed for more than 55% of COVID-19 outpatients. The three most commonly prescribed antibiotics were azithromycin, amoxicillin/clavulanic acid, and doxycycline. The trend of repurposing antibiotics for COVID-19 and other diseases treatment might be a double-edged sword. The long-term effect of this practice might be an increase in antimicrobial resistance and a loss of antibiotic effectiveness.