Promises and pitfalls of electronic health record analysis

Equity in Mental Health Care Receipt among Youth Who Use School-Based Health Centers

INTRODUCTION

Youth experience significant mental health (MH) needs, and gender- and racially/ethnically-diverse youth are less likely than peers to receive care. School-based health centers (SBHCs) are a healthcare delivery model that may decrease disparities. This study examined the role of SBHCs in reducing disparities in MH care receipt among SBHC clients.

METHODS

Data from electronic health records of 5,396 youth ages 12 to 21 years who visited 14 SBHCs in one California county from 2021 to 2023 were analyzed in 2023-2024 using multiple logistic regression to assess disparities in MH care receipt and depression screenings.

RESULTS

Receipt of MH care from SBHCs varied significantly by gender but not age, sexual orientation, or race/ethnicity. Compared to female clients, males had reduced odds (AOR: 0.50) and gender-diverse clients had higher odds (AOR: 2.70) of receiving MH care. For receipt of depression screenings, male clients had reduced odds (AOR: 0.86); Latino clients had higher odds than white clients (AOR: 1.80); and older adolescents and young adults had higher odds than younger adolescents (AORs: 1.44 and 1.45, respectively). Receipt of follow-up MH care after a positive depression result varied only by gender, with male clients having reduced odds (AOR: 0.63).

DISCUSSION

SBHCs may reach youth who are traditionally less likely to seek care in other settings, including racially/ethnically- and gender-diverse youth. As in other settings, engaging males in healthcare is an area for improvement. These findings help to demonstrate the potential of SBHCs for decreasing disparities in mental health care.

The role and limitations of electronic medical records versus patient interviews for determining symptoms of, underlying comorbidities of, and medication use by patients with COVID-19

Electronic medical records (EMRs) are important for rapidly compiling information to determine disease characteristics (eg, symptoms) and risk factors (eg, underlying comorbidities, medications) for disease-related outcomes. To assess EMR data accuracy, agreement between EMR abstractions and patient interviews was evaluated. Symptoms, medical history, and medication use among patients with COVID-19 collected from EMRs and patient interviews were compared using overall agreement (ie, same answer in EMR and interview), reported agreement (yes answer in both EMR and interview among those who reported yes in either), and κ statistics. Overall, patients reported more symptoms in interviews than in EMR abstractions. Overall agreement was high (≥50% for 20 of 23 symptoms), but only subjective fever and dyspnea had reported agreement of ≥50%. The κ statistics for symptoms were generally low. Reported medical conditions had greater agreement with all condition categories (n = 10 of 10) having ≥50% overall agreement and half (n = 5 of 10) having ≥50% reported agreement. More nonprescription medications were reported in interviews than in EMR abstractions, leading to low reported agreement (28%). Discordance was observed for symptoms, medical history, and medication use between EMR abstractions and patient interviews. Investigations using EMRs to describe clinical characteristics and identify risk factors should consider the potential for incomplete data, particularly for symptoms and medications.

Checklist and guidance on creating codelists for routinely collected health data research

Background

Codelists are required to extract meaningful information on characteristics and events from routinely collected health data such as electronic health records. Research using routinely collected health data relies on codelists to define study populations and variables, thus, trustworthy codelists are important. Here, we provide a checklist, in the style of commonly used reporting guidelines, to help researchers adhere to best practice in codelist development and sharing.

Methods

Based on a literature search and a workshop with researchers experienced in the use of routinely collected health data, we created a set of recommendations that are 1. broadly applicable to different datasets, research questions, and methods of codelist creation; 2. easy to follow, implement and document by an individual researcher, and 3. fit within a step-by-step process. We then formatted these recommendations into a checklist.

Results

We have created a 10-step checklist, comprising 28 items, with accompanying guidance on each step. The checklist advises on which metadata to provide, how to define a clinical concept, how to identify and evaluate existing codelists, how to create new codelists, and how to review, check, finalise, and publish a created codelist.

Conclusions

Use of the checklist can reassure researchers that best practice was followed during the development of their codelists, increasing trust in research that relies on these codelists and facilitating wider re-use and adaptation by other researchers.

Usability evaluation of electronic health records at the trauma and emergency directorates at the Komfo Anokye teaching hospital in the Ashanti region of Ghana

BACKGROUND

Electronic health records (EHRs) are currently gaining popularity in emerging economies because they provide options for exchanging patient data, increasing operational efficiency, and improving patient outcomes. This study examines how service providers at Ghana's Komfo Anokye Teaching Hospital adopt and use an electronic health records (EHRs) system. The emphasis is on identifying factors impacting adoption and the problems that healthcare personnel encounter in efficiently using the EHRs system.

METHOD

A quantitative cross-sectional technique was utilised to collect data from 234 trauma and emergency department staff members via standardised questionnaires. The participants were selected using the purposive sampling method. The Pearson Chi-square Test was used to examine the relationship between respondents' acceptability and use of EHRs.

RESULTS

The study discovered that a sizable number of respondents (86.8%) embraced and actively used the EHRs system. However, other issues were noted, including insufficient system training and malfunctions (35.9%), power outages (18.8%), privacy concerns (9.4%), and insufficient maintenance (4.7%). The respondents' comfortability in using the electronic health record system (X2=11.30, p=0.001), system dependability (X2=30.74, p=0.0001), and EHR's ability to reduce patient waiting time (X2=14.39, p=0.0001) were all strongly associated with their degree of satisfaction with the system. Furthermore, respondents who said elects increase patient care (X2= 75.59, p = 0.0001) and income creation (X2= 8.48, p = 0.004), which is related to the acceptability of the electronic health records system.

CONCLUSION

The study revealed that comfort, reliability, and improved care quality all had an impact on the EHRs system's acceptability and utilization. Challenges, including equipment malfunctions and power outages, were found. Continuous professional training was emphasized as a means of increasing employee confidence, as did the construction of a power backup system to combat disruptions. Patient data privacy was highlighted. In conclusion, this study highlights the relevance of EHRs system adoption and usability in healthcare. While the benefits are obvious, addressing obstacles through training, technical support, and infrastructure improvements is critical for increasing system effectiveness.

Clinical coding of long COVID in primary care 2020-2023 in a cohort of 19 million adults: an OpenSAFELY analysis

Background

Long COVID is the patient-coined term for the persistent symptoms of COVID-19 illness for weeks, months or years following the acute infection. There is a large burden of long COVID globally from self-reported data, but the epidemiology, causes and treatments remain poorly understood. Primary care is used to help identify and treat patients with long COVID and therefore Electronic Health Records (EHRs) of past COVID-19 patients could be used to help fill these knowledge gaps. We aimed to describe the incidence and differences in demographic and clinical characteristics in recorded long COVID in primary care records in England.

Methods

With the approval of NHS England we used routine clinical data from over 19 million adults in England linked to SARS-COV-2 test result, hospitalisation and vaccination data to describe trends in the recording of 16 clinical codes related to long COVID between November 2020 and January 2023. Using OpenSAFELY, we calculated rates per 100,000 person-years and plotted how these changed over time. We compared crude and adjusted (for age, sex, 9 NHS regions of England, and the dominant variant circulating) rates of recorded long COVID in patient records between different key demographic and vaccination characteristics using negative binomial models.

Findings

We identified a total of 55,465 people recorded to have long COVID over the study period, which included 20,025 diagnoses codes and 35,440 codes for further assessment. The incidence of new long COVID records increased steadily over 2021, and declined over 2022. The overall rate per 100,000 person-years was 177.5 cases in women (95% CI: 175.5-179) and 100.5 in men (99.5-102). The majority of those with a long COVID record did not have a recorded positive SARS-COV-2 test 12 or more weeks before the long COVID record.

Interpretation

In this descriptive study, EHR recorded long COVID was very low between 2020 and 2023, and incident records of long COVID declined over 2022. Using EHR diagnostic or referral codes unfortunately has major limitations in identifying and ascertaining true cases and timing of long COVID.

Funding

This research was supported by the National Institute for Health and Care Research (NIHR) (OpenPROMPT: COV-LT2-0073).

AssistMED project: Transforming cardiology cohort characterisation from electronic health records through natural language processing - Algorithm design, preliminary results, and field prospects

INTRODUCTION

Electronic health records (EHR) are of great value for clinical research. However, EHR consists primarily of unstructured text which must be analysed by a human and coded into a database before data analysis- a time-consuming and costly process limiting research efficiency. Natural language processing (NLP) can facilitate data retrieval from unstructured text. During AssistMED project, we developed a practical, NLP tool that automatically provides comprehensive clinical characteristics of patients from EHR, that is tailored to clinical researchers needs.

MATERIAL AND METHODS

AssistMED retrieves patient characteristics regarding clinical conditions, medications with dosage, and echocardiographic parameters with clinically oriented data structure and provides researcher-friendly database output. We validate the algorithm performance against manual data retrieval and provide critical quantitative and qualitative analysis.

RESULTS

AssistMED analysed the presence of 56 clinical conditions, medications from 16 drug groups with dosage and 15 numeric echocardiographic parameters in a sample of 400 patients hospitalized in the cardiology unit. No statistically significant differences between algorithm and human retrieval were noted. Qualitative analysis revealed that disagreements with manual annotation were primarily accounted to random algorithm errors, erroneous human annotation and lack of advanced context awareness of our tool.

CONCLUSIONS

Current NLP approaches are feasible to acquire accurate and detailed patient characteristics tailored to clinical researchers' needs from EHR. We present an in-depth description of an algorithm development and validation process, discuss obstacles and pinpoint potential solutions, including opportunities arising with recent advancements in the field of NLP, such as large language models.

Temporal Risk of Nonfatal Cardiovascular Events After Chronic Obstructive Pulmonary Disease Exacerbation: A Population-based Study

Rationale: Cardiovascular events after chronic obstructive pulmonary disease (COPD) exacerbations are recognized. Studies to date have been post hoc analyses of trials, did not differentiate exacerbation severity, included death in the cardiovascular outcome, or had insufficient power to explore individual outcomes temporally.Objectives: We explore temporal relationships between moderate and severe exacerbations and incident, nonfatal hospitalized cardiovascular events in a primary care-derived COPD cohort.Methods: We included people with COPD in England from 2014 to 2020, from the Clinical Practice Research Datalink Aurum primary care database. The index date was the date of first COPD exacerbation or, for those without exacerbations, date upon eligibility. We determined composite and individual cardiovascular events (acute coronary syndrome, arrhythmia, heart failure, ischemic stroke, and pulmonary hypertension) from linked hospital data. Adjusted Cox regression models were used to estimate average and time-stratified adjusted hazard ratios (aHRs).Measurements and Main Results: Among 213,466 patients, 146,448 (68.6%) had any exacerbation; 119,124 (55.8%) had moderate exacerbations, and 27,324 (12.8%) had severe exacerbations. A total of 40,773 cardiovascular events were recorded. There was an immediate period of cardiovascular relative rate after any exacerbation (1-14 d; aHR, 3.19 [95% confidence interval (CI), 2.71-3.76]), followed by progressively declining yet maintained effects, elevated after one year (aHR, 1.84 [95% CI, 1.78-1.91]). Hazard ratios were highest 1-14 days after severe exacerbations (aHR, 14.5 [95% CI, 12.2-17.3]) but highest 14-30 days after moderate exacerbations (aHR, 1.94 [95% CI, 1.63-2.31]). Cardiovascular outcomes with the greatest two-week effects after a severe exacerbation were arrhythmia (aHR, 12.7 [95% CI, 10.3-15.7]) and heart failure (aHR, 8.31 [95% CI, 6.79-10.2]).Conclusions: Cardiovascular events after moderate COPD exacerbations occur slightly later than after severe exacerbations; heightened relative rates remain beyond one year irrespective of severity. The period immediately after an exacerbation presents a critical opportunity for clinical intervention and treatment optimization to prevent future cardiovascular events.

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.

Factors Associated With Treatment Escalation for Psoriasis: An Analysis of Electronic Health Records Data

Background

Electronic health records (EHRs) offer the possibility of using data entry templates to simultaneously document routine clinical care and capture disease-specific measures as discrete data elements that can be used for health services research (HSR). The objective of this study was to determine factors associated with meaningful treatment escalation (MTE) of psoriasis as a pilot study for future real-world HSR studies.

Methods

We conducted a retrospective, observational cohort study of psoriasis patients by using data collected during routine clinical care from an EHR using EpiCare® SmartForms. The psoriasis SmartForm records psoriasis disease severity measures and descriptive findings to generate visit notes. These data were extracted and analyzed to identify factors associated with MTE, defined as changing or adding, phototherapy, systemic, or biologic therapy.

Results

473 psoriasis patients met study criteria; 239 underwent MTE between their first and third observed visits. Patients who experienced MTE had more severe disease at Visit 1-assessed by BSA, pPGA, oPGA, and a patient-reported disease severity measure--than patients who did not experience MTE. Other factors associated with MTE included use of topicals only or no active treatment at Visit 1, palmoplantar disease, and involvement of other difficult-to-treat body areas. Patients who underwent MTE experienced larger improvements in disease severity than those who did not.

Conclusions

This study highlights how data collected during routine clinical practice can be readily used for real-world retrospective HSR when disease measures are captured as discrete elements. This approach could provide a cost-effective platform to conduct real-world HSR.

Electronic health records (EHRs) in clinical research and platform trials: Application of the innovative EHR-based methods developed by EU-PEARL

OBJECTIVE

Electronic Health Record (EHR) systems are digital platforms in clinical practice used to collect patients' clinical information related to their health status and represents a useful storage of real-world data. EHRs have a potential role in research studies, in particular, in platform trials. Platform trials are innovative trial designs including multiple trial arms (conducted simultaneously and/or sequentially) on different treatments under a single master protocol. However, the use of EHRs in research comes with important challenges such as incompleteness of records and the need to translate trial eligibility criteria into interoperable queries. In this paper, we aim to review and to describe our proposed innovative methods to tackle some of the most important challenges identified. This work is part of the Innovative Medicines Initiative (IMI) EU Patient-cEntric clinicAl tRial pLatforms (EU-PEARL) project's work package 3 (WP3), whose objective is to deliver tools and guidance for EHR-based protocol feasibility assessment, clinical site selection, and patient pre-screening in platform trials, investing in the building of a data-driven clinical network framework that can execute these complex innovative designs for which feasibility assessments are critically important.

METHODS

ISO standards and relevant references informed a readiness survey, producing 354 criteria with corresponding questions selected and harmonised through a 7-round scoring process (0-1) in stakeholder meetings, with 85% of consensus being the threshold of acceptance for a criterium/question. ATLAS cohort definition and Cohort Diagnostics were mainly used to create the trial feasibility eligibility (I/E) criteria as executable interoperable queries.

RESULTS

The WP3/EU-PEARL group developed a readiness survey (eSurvey) for an efficient selection of clinical sites with suitable EHRs, consisting of yes-or-no questions, and a set-up of interoperable proxy queries using physicians' defined trial criteria. Both actions facilitate recruiting trial participants and alignment between study costs/timelines and data-driven recruitment potential.

CONCLUSION

The eSurvey will help create an archive of clinical sites with mature EHR systems suitable to participate in clinical trials/platform trials, and the interoperable proxy queries of trial eligibility criteria will help identify the number of potential participants. Ultimately, these tools will contribute to the production of EHR-based protocol design.

Potential effect of antidepressants on remission from cocaine use disorder - A nationwide matched retrospective cohort study

BACKGROUND

Cocaine use disorder (CUD) is a significant public health issue for which there is no Food and Drug Administration-approved pharmacotherapy. Depressive disorders are common psychiatric comorbidity amongst individuals with CUD.

METHODS

A retrospective cohort study was conducted among 161,544 patients diagnosed with CUD and depression to evaluate the effectiveness of 13 antidepressants on CUD remission. For any antidepressant found to be associated with CUD remission that had an additional indication, we conducted an additional analysis to evaluate the effectiveness of the candidate drug in patients with CUD with that indication. We then analyzed publicly genomic and functional databases to identify potential explanatory mechanisms of action of the candidate drug in the treatment of CUD.

RESULTS

Among these antidepressants, bupropion was associated with higher rates of CUD remission compared to propensity-score matched patients prescribed other antidepressants: hazard ratio (HR) and 95% confidence interval (CI) 1.57 (95% CI: 1.27-1.94). Bupropion is also approved for smoking cessation. We identified CUD patients with co-occurring nicotine dependence and observed that patients prescribed bupropion displayed a higher rate of CUD remission compared to matched individuals prescribed other drugs for nicotine dependence: 1.38 (95% CI: 1.11-1.71). Genetic and functional analyses revealed that bupropion interacts with four protein-encoding genes (COMT, DRD2, SLC6A3, and SLC6A4) which are also associated with CUD and targets CUD-associated pathways including serotonergic synapses, cocaine addiction, and dopaminergic synapses.

CONCLUSIONS

Our findings suggest that bupropion might be considered a treatment for improving CUD remission in patients with CUD and co-occurring depression or nicotine dependence.

Challenges and Opportunities for Data Science in Women's Health

The intersection of women's health and data science is a field of research that has historically trailed other fields, but more recently it has gained momentum. This growth is being driven not only by new investigators who are moving into this area but also by the significant opportunities that have emerged in new methodologies, resources, and technologies in data science. Here, we describe some of the resources and methods being used by women's health researchers today to meet challenges in biomedical data science. We also describe the opportunities and limitations of applying these approaches to advance women's health outcomes and the future of the field, with emphasis on repurposing existing methodologies for women's health.

Living alone and mental health: parallel analyses in UK longitudinal population surveys and electronic health records prior to and during the COVID-19 pandemic

BACKGROUND

People who live alone experience greater levels of mental illness; however, it is unclear whether the COVID-19 pandemic had a disproportionately negative impact on this demographic.

OBJECTIVE

To describe the mental health gap between those who live alone and with others in the UK prior to and during the COVID-19 pandemic.

METHODS

Self-reported psychological distress and life satisfaction in 10 prospective longitudinal population surveys (LPSs) assessed in the nearest pre-pandemic sweep and three periods during the pandemic. Recorded diagnosis of common and severe mental illnesses between March 2018 and January 2022 in electronic healthcare records (EHRs) within the OpenSAFELY-TPP.

FINDINGS

In 37 544 LPS participants, pooled models showed greater psychological distress (standardised mean difference (SMD): 0.09 (95% CI: 0.04; 0.14); relative risk: 1.25 (95% CI: 1.12; 1.39)) and lower life satisfaction (SMD: -0.22 (95% CI: -0.30; -0.15)) for those living alone pre-pandemic. This gap did not change during the pandemic. In the EHR analysis of c.16 million records, mental health conditions were more common in those who lived alone (eg, depression 26 (95% CI: 18 to 33) and severe mental illness 58 (95% CI: 54 to 62) more cases more per 100 000). For common mental health disorders, the gap in recorded cases in EHRs narrowed during the pandemic.

CONCLUSIONS

People living alone have poorer mental health and lower life satisfaction. During the pandemic, this gap in self-reported distress remained; however, there was a narrowing of the gap in service use.

CLINICAL IMPLICATIONS

Greater mental health need and potentially greater barriers to mental healthcare access for those who live alone need to be considered in healthcare planning.

A network medicine approach to study comorbidities in heart failure with preserved ejection fraction

BACKGROUND

Comorbidities are expected to impact the pathophysiology of heart failure (HF) with preserved ejection fraction (HFpEF). However, comorbidity profiles are usually reduced to a few comorbid disorders. Systems medicine approaches can model phenome-wide comorbidity profiles to improve our understanding of HFpEF and infer associated genetic profiles.

METHODS

We retrospectively explored 569 comorbidities in 29,047 HF patients, including 8062 HFpEF and 6585 HF with reduced ejection fraction (HFrEF) patients from a German university hospital. We assessed differences in comorbidity profiles between HF subtypes via multiple correspondence analysis. Then, we used machine learning classifiers to identify distinctive comorbidity profiles of HFpEF and HFrEF patients. Moreover, we built a comorbidity network (HFnet) to identify the main disease clusters that summarized the phenome-wide comorbidity. Lastly, we predicted novel gene candidates for HFpEF by linking the HFnet to a multilayer gene network, integrating multiple databases. To corroborate HFpEF candidate genes, we collected transcriptomic data in a murine HFpEF model. We compared predicted genes with the murine disease signature as well as with the literature.

RESULTS

We found a high degree of variance between the comorbidity profiles of HFpEF and HFrEF, while each was more similar to HFmrEF. The comorbidities present in HFpEF patients were more diverse than those in HFrEF and included neoplastic, osteologic and rheumatoid disorders. Disease communities in the HFnet captured important comorbidity concepts of HF patients which could be assigned to HF subtypes, age groups, and sex. Based on the HFpEF comorbidity profile, we predicted and recovered gene candidates, including genes involved in fibrosis (COL3A1, LOX, SMAD9, PTHL), hypertrophy (GATA5, MYH7), oxidative stress (NOS1, GSST1, XDH), and endoplasmic reticulum stress (ATF6). Finally, predicted genes were significantly overrepresented in the murine transcriptomic disease signature providing additional plausibility for their relevance.

CONCLUSIONS

We applied systems medicine concepts to analyze comorbidity profiles in a HF patient cohort. We were able to identify disease clusters that helped to characterize HF patients. We derived a distinct comorbidity profile for HFpEF, which was leveraged to suggest novel candidate genes via network propagation. The identification of distinctive comorbidity profiles and candidate genes from routine clinical data provides insights that may be leveraged to improve diagnosis and identify treatment targets for HFpEF patients.

Opioid dose and pain effects of an online pain self-management program to augment usual care in adults with chronic pain: a multisite randomized clinical trial

ABSTRACT

Readily accessible nonpharmacological interventions that can assist in opioid dose reduction while managing pain is a priority for adults receiving long-term opioid therapy (LOT). Few large-scale evaluations of online pain self-management programs exist that capture effects on reducing morphine equivalent dose (MED) simultaneously with pain outcomes. An open-label, intent-to-treat, randomized clinical trial recruited adults (n = 402) with mixed chronic pain conditions from primary care and pain clinics of 2 U.S. academic healthcare systems. All participants received LOT-prescriber-provided treatment of MED ≥ 20 mg while receiving either E-health (a 4-month subscription to the online Goalistics Chronic Pain Management Program), or treatment as usual (TAU). Among 402 participants (279 women [69.4%]; mean [SD] age, 56.7 [11.0] years), 200 were randomized to E-health and 202 to TAU. Of 196 E-heath participants, 105 (53.6%) achieved a ≥15% reduction in daily MED compared with 85 (42.3%) of 201 TAU participants (odds ratio, 1.6 [95% CI, 1.1-2.3]; P = 0.02); number-needed-to-treat was 8.9 (95% CI, 4.8, 66.0). Of 166 E-health participants, 24 (14.5%) achieved a ≥2 point decrease in pain intensity vs 13 (6.8%) of 192 TAU participants (odds ratio, 2.4 [95% CI, 1.2-4.9]; P = 0.02). Benefits were also observed in pain knowledge, pain self-efficacy, and pain coping. The findings suggest that for adults on LOT for chronic pain, use of E-health, compared with TAU, significantly increased participants' likelihood of clinically meaningful decreases in MED and pain. This low-burden online intervention could assist adults on LOT in reducing daily opioid use while self-managing pain symptom burdens.

Characterising treatment outcomes of patients achieving quarterly aflibercept dosing for neovascular age-related macular degeneration: real-world clinical outcomes from a large tertiary care centre

BACKGROUND AND OBJECTIVE

To evaluate the proportion of patients achieving a 12-week (q12) aflibercept dosing interval in patients with neovascular age-related macular degeneration (nAMD).

PATIENTS AND METHODS

Retrospective, comparative, non-randomised electronic medical record (EMR) database study of the Moorfields database of treatment-naïve nAMD eyes. Extraction criteria included at least 7 aflibercept injections in first year of treatment, AMD in the diagnosis field of EMR, and minimum of 1 year follow-up data.

RESULTS

There were 2416 eyes of 2163 patients started on anti-vascular endothelial growth factor (anti-VEGF) between 01-11-2013 & 14-02-2020 who had received at least 7 aflibercept intravitreal injections (electronic database accessed March 2021). Of these, 1674 (68%) eyes of 1537 patients had at least one q12 dosing interval (>=84 and < =98 days between injections) during the first 2 years of treatment. This included 926 (61.8%) female patients and 856 (right eyes age at 1st injection), 936 (62.4%) Caucasian, and 32 (2.1%) Afro-Caribbean patients. The median time to the first q12 injection (95% confidence interval) was 1.76 years (1.70-1.86) with mean (±SD) of 11.8 (±6.0) injections. Visual acuity (ETDRS letters) of the eyes without q12 injection and eyes with a q12 injection was 57.9 ± 14.7 and 56.7 ± 14.8 respectively at baseline, 61.4 ± 18.1 and 63.0 ± 15.9 respectively at 12 months and 61.2 ± 20.1 and 61.1 ± 17.8 respectively at 24 months.

CONCLUSION

68% of eyes were able to achieve a q12 injection dose within the first 2 years of treatment. Eyes achieving a q12 injection in the first 2 years achieved a similar visual acuity outcome at both 1 and 2-year follow-up to those unable to do so, with a fewer number of total injections.

Informative missingness: What can we learn from patterns in missing laboratory data in the electronic health record?

BACKGROUND

In electronic health records, patterns of missing laboratory test results could capture patients' course of disease as well as ​​reflect clinician's concerns or worries for possible conditions. These patterns are often understudied and overlooked. This study aims to identify informative patterns of missingness among laboratory data collected across 15 healthcare system sites in three countries for COVID-19 inpatients.

METHODS

We collected and analyzed demographic, diagnosis, and laboratory data for 69,939 patients with positive COVID-19 PCR tests across three countries from 1 January 2020 through 30 September 2021. We analyzed missing laboratory measurements across sites, missingness stratification by demographic variables, temporal trends of missingness, correlations between labs based on missingness indicators over time, and clustering of groups of labs based on their missingness/ordering pattern.

RESULTS

With these analyses, we identified mapping issues faced in seven out of 15 sites. We also identified nuances in data collection and variable definition for the various sites. Temporal trend analyses may support the use of laboratory test result missingness patterns in identifying severe COVID-19 patients. Lastly, using missingness patterns, we determined relationships between various labs that reflect clinical behaviors.

CONCLUSION

In this work, we use computational approaches to relate missingness patterns to hospital treatment capacity and highlight the heterogeneity of looking at COVID-19 over time and at multiple sites, where there might be different phases, policies, etc. Changes in missingness could suggest a change in a patient's condition, and patterns of missingness among laboratory measurements could potentially identify clinical outcomes. This allows sites to consider missing data as informative to analyses and help researchers identify which sites are better poised to study particular questions.

Machine Learning Models Using Routinely Collected Clinical Data Offer Robust and Interpretable Predictions of 90-Day Unplanned Acute Care Use for Cancer Immunotherapy Patients

PURPOSE

Clinical management of patients receiving immune checkpoint inhibitors (ICIs) could be informed using accurate predictive tools to identify patients at risk of short-term acute care utilization (ACU). We used routinely collected data to develop and assess machine learning (ML) algorithms to predict unplanned ACU within 90 days of ICI treatment initiation.

METHODS

We used aggregated electronic health record data from 7,960 patients receiving ICI treatments to train and assess eight ML algorithms. We developed the models using pre-SARS-COV-19 COVID-19 data generated between January 2016 and February 2020. We validated our algorithms using data collected between March 2020 and June 2022 (peri-COVID-19 sample). We assessed performance using area under the receiver operating characteristic curves (AUROC), sensitivity, specificity, and calibration plots. We derived intuitive explanations of predictions using variable importance and Shapley additive explanation analyses. We assessed the marginal performance of ML models compared with that of univariate and multivariate logistic regression (LR) models.

RESULTS

Most algorithms significantly outperformed the univariate and multivariate LR models. The extreme gradient boosting trees (XGBT) algorithm demonstrated the best overall performance (AUROC, 0.70; sensitivity, 0.53; specificity, 0.74) on the peri-COVID-19 sample. The algorithm performance was stable across both pre- and peri-COVID-19 samples, as well as ICI regimen and cancer groups. Type of ICI agents, oxygen saturation, diastolic blood pressure, albumin level, platelet count, immature granulocytes, absolute monocyte, chloride level, red cell distribution width, and alcohol intake were the top 10 key predictors used by the XGBT algorithm.

CONCLUSION

Machine learning algorithms trained using routinely collected data outperformed traditional statistical models when predicting 90-day ACU. The XGBT algorithm has the potential to identify high-ACU risk patients and enable preventive interventions to avoid ACU.

Ethnic differences in hypertension management, medication use and blood pressure control in UK primary care, 2006-2019: a retrospective cohort study

Background

In the UK, previous work suggests ethnic inequalities in hypertension management. We studied ethnic differences in hypertension management and their contribution to blood pressure (BP) control.

Methods

We conducted a cohort study of antihypertensive-naïve individuals of European, South Asian and African/African Caribbean ethnicity with a new raised BP reading in UK primary care from 2006 to 2019, using the Clinical Practice Research Datalink (CPRD). We studied differences in: BP re-measurement after an initial hypertensive BP, antihypertensive initiation, BP monitoring, antihypertensive intensification, antihypertensive persistence/adherence and BP control one year after antihypertensive initiation. Models adjusted for socio-demographics, BP, comorbidity, healthcare usage and polypharmacy (plus antihypertensive class, BP monitoring, intensification, persistence and adherence for BP control models).

Findings

A total of 731,506 (93.5%), 30,379 (3.9%) and 20,256 (2.6%) people of European, South Asian and African/African Caribbean ethnicity were studied. Hypertension management indicators were similar or more favourable for South Asian than European groups (OR/HR [95% CI] in fully-adjusted models of BP re-measurement: 1.16 [1.09, 1.24]), antihypertensive initiation: 1.49 [1.37, 1.62], BP monitoring: 0.97 [0.94, 1.00] and antihypertensive intensification: 1.10 [1.04, 1.16]). For people of African/African Caribbean ethnicity, BP re-measurement rates were similar to those of European ethnicity (0.98 [0.91, 1.05]), and antihypertensive initiation rates greater (1.48 [1.32, 1.66]), but BP monitoring (0.91 [0.87, 0.95]) and intensification rates lower (0.93 [0.87, 1.00]). Persistence and adherence were lower in South Asian (0.48 [0.45, 0.51] and 0.51 [0.47, 0.56]) and African/African Caribbean (0.38 [0.35, 0.42] and 0.39 [0.36, 0.43]) than European groups. BP control was similar in South Asian and less likely in African/African Caribbean than European groups (0.98 [0.90, 1.06] and 0.81 [0.74, 0.89] in age, gender and BP adjusted models). The latter difference attenuated after adjustment for persistence (0.91 [0.82, 0.99]) or adherence (0.92 [0.83, 1.01]), and was absent for antihypertensive-adherent people (0.99 [0.88, 1.10]).

Interpretation

We demonstrate that antihypertensive initiation does not vary by ethnicity, but subsequent BP control was notably lower among people of African/African Caribbean ethnicity, potentially associated with being less likely to remain on regular treatment. A nationwide strategy to understand and address differences in ongoing management of people on antihypertensives is imperative.

Funding

Diabetes UK.

Development and validation of algorithms to build an electronic health record based cohort of patients with systemic sclerosis

OBJECTIVES

To evaluate methods of identifying patients with systemic sclerosis (SSc) using International Classification of Diseases, Tenth Revision (ICD-10) codes (M34*), electronic health record (EHR) databases and organ involvement keywords, that result in a validated cohort comprised of true cases with high disease burden.

METHODS

We retrospectively studied patients in a healthcare system likely to have SSc. Using structured EHR data from January 2016 to June 2021, we identified 955 adult patients with M34* documented 2 or more times during the study period. A random subset of 100 patients was selected to validate the ICD-10 code for its positive predictive value (PPV). The dataset was then divided into a training and validation sets for unstructured text processing (UTP) search algorithms, two of which were created using keywords for Raynaud's syndrome, and esophageal involvement/symptoms.

RESULTS

Among 955 patients, the average age was 60. Most patients (84%) were female; 75% of patients were White, and 5.2% were Black. There were approximately 175 patients per year with the code newly documented, overall 24% had an ICD-10 code for esophageal disease, and 13.4% for pulmonary hypertension. The baseline PPV was 78%, which improved to 84% with UTP, identifying 788 patients likely to have SSc. After the ICD-10 code was placed, 63% of patients had a rheumatology office visit. Patients identified by the UTP search algorithm were more likely to have increased healthcare utilization (ICD-10 codes 4 or more times 84.1% vs 61.7%, p < .001), organ involvement (pulmonary hypertension 12.7% vs 6% p = .011) and medication use (mycophenolate use 28.7% vs 11.4%, p < .001) than those identified by the ICD codes alone.

CONCLUSION

EHRs can be used to identify patients with SSc. Using unstructured text processing keyword searches for SSc clinical manifestations improved the PPV of ICD-10 codes alone and identified a group of patients most likely to have SSc and increased healthcare needs.

Improving Cohort-Hospital Matching Accuracy through Standardization and Validation of Participant Identifiable Information

Linking very large, consented birth cohorts to birthing hospitals clinical data could elucidate the lifecourse outcomes of health care and exposures during the pregnancy, birth and newborn periods. Unfortunately, cohort personally identifiable information (PII) often does not include unique identifier numbers, presenting matching challenges. To develop optimized cohort matching to birthing hospital clinical records, this pilot drew on a one-year (December 2020-December 2021) cohort for a single Australian birthing hospital participating in the whole-of-state Generation Victoria (GenV) study. For 1819 consented mother-baby pairs and 58 additional babies (whose mothers were not themselves participating), we tested the accuracy and effort of various approaches to matching. We selected demographic variables drawn from names, DOB, sex, telephone, address (and birth order for multiple births). After variable standardization and validation, accuracy rose from 10% to 99% using a deterministic-rule-based approach in 10 steps. Using cohort-specific modifications of the Australian Statistical Linkage Key (SLK-581), it took only 3 steps to reach 97% (SLK-5881) and 98% (SLK-5881.1) accuracy. We conclude that our SLK-5881 process could safely and efficiently achieve high accuracy at the population level for future birth cohort-birth hospital matching in the absence of unique identifier numbers.

Evaluating novel methods of outcome assessment following cardiac arrest

INTRODUCTION

We compared novel methods of long-term follow-up after resuscitation from cardiac arrest to a query of the National Death Index (NDI). We hypothesized use of the electronic health record (EHR), and internet-based sources would have high sensitivity for identifying decedents identified by the NDI.

METHODS

We performed a retrospective study including patients treated after cardiac arrest at a single academic center from 2010 to 2018. We evaluated two novel methods to ascertain long-term survival and modified Rankin Scale (mRS): 1) a structured chart review of our health system's EHR; and 2) an internet-based search of: a) local newspapers, b) Ancestry.com, c) Facebook, d) Twitter, e) Instagram, and f) Google. If a patient was not reported deceased by any source, we considered them to be alive. We compared results of these novel methods to the NDI to calculate sensitivity. We queried the NDI for 200 in-hospital decedents to evaluate sensitivity against a true criterion standard.

RESULTS

We included 1,097 patients, 897 (82%) alive at discharge and 200 known decedents (18%). NDI identified 197/200 (99%) of known decedents. The EHR and local newspapers had highest sensitivity compared to the NDI (87% and 86% sensitivity, respectively). Online sources identified 10 likely decedents not identified by the NDI. Functional status estimated from EHR, and internet sources at follow up agreed in 38% of alive patients.

CONCLUSIONS

Novel methods of outcome assessment are an alternative to NDI for determining patients' vital status. These methods are less reliable for estimating functional status.

Examining Mental Health Disorders in Overweight and Obese Pediatric Patients

INTRODUCTION

We investigated the frequency and variation in three mental health diagnoses among obese or overweight children and adolescents.

METHOD

Logistic regression was used to examine the association between the outcome variables-anxiety, depression, and adjustment disorders-with the following covariates: overweight/obesity status, sex, age, and race.

RESULTS

Findings show anxiety, depressive, and adjustment disorder diagnoses were significantly higher for overweight or obese youth in our sample. In addition, diagnosis rates for one or more of these disorders increase as children grow into adolescence. Furthermore, we found significantly higher rates of depression and significantly lower rates of anxiety among youth who live in places with higher rates of poverty.

DISCUSSION

Findings indicate a target age for providers to focus on mental health screening among overweight/obese patients: (1) early adolescence (aged 11-14 years) for depressive and adjustment disorders and (2) early childhood (aged 2-4 years) for anxiety disorder.

Rehabilitation services and related health databases, Japan

The demographic transition towards an ageing population and the epidemiological transition from communicable to noncommunicable diseases have increased the demand for rehabilitation services globally. The aims of this paper were to describe the integration of rehabilitation into the Japanese health system and to illustrate how health information systems containing real-world data can be used to improve rehabilitation services, especially for the ageing population of Japan. In addition, there is an overview of how evidence-informed rehabilitation policy is guided by the analysis of large Japanese health databases, such as: (i) the National Database of Health Insurance Claims and Specific Health Checkups; (ii) the long-term care insurance comprehensive database; and (iii) the Long-Term Care Information System for Evidence database. Especially since the 1990s, the integration of rehabilitation into the Japanese health system has been driven by the country’s ageing population and rehabilitation is today provided widely to an increasing number of older adults. General medical insurance in Japan covers acute and post-acute (or recovery) intensive rehabilitation. Long-term care insurance covers rehabilitation at long-term care institutions and community facilities for older adults with the goal of helping to maintain independence in an ageing population. The analysis of large health databases can be used to improve the management of rehabilitation care services and increase scientific knowledge as well as guide rehabilitation policy and practice. In particular, such analyses could help solve the current challenges of overtreatment and undertreatment by identifying strict criteria for determining who should receive long-term rehabilitation services.

ExPRSweb: An online repository with polygenic risk scores for common health-related exposures

Complex traits are influenced by genetic risk factors, lifestyle, and environmental variables, so-called exposures. Some exposures, e.g., smoking or lipid levels, have common genetic modifiers identified in genome-wide association studies. Because measurements are often unfeasible, exposure polygenic risk scores (ExPRSs) offer an alternative to study the influence of exposures on various phenotypes. Here, we collected publicly available summary statistics for 28 exposures and applied four common PRS methods to generate ExPRSs in two large biobanks: the Michigan Genomics Initiative and the UK Biobank. We established ExPRSs for 27 exposures and demonstrated their applicability in phenome-wide association studies and as predictors for common chronic conditions. Especially the addition of multiple ExPRSs showed, for several chronic conditions, an improvement compared to prediction models that only included traditional, disease-focused PRSs. To facilitate follow-up studies, we share all ExPRS constructs and generated results via an online repository called ExPRSweb.

Economic burden in US patients with newly diagnosed acute myeloid leukemia receiving intensive induction chemotherapy

Aim: This retrospective, observational study assessed healthcare resource utilization (HCRU) and costs for newly diagnosed acute myeloid leukemia (AML) patients receiving intensive induction chemotherapy. Materials & methods: Adult AML patients with inpatient hospitalization or hospital-based outpatient visit receiving intensive induction chemotherapy (CPX-351 or 7 + 3 treatments) were identified from the Premier Healthcare Database (US). Results: All 642 patients had inpatient hospitalizations (median number = 2; median length of stay = 16 days); 22.4% had an ICU admission. Median total outpatient hospital cost was US$2904 per patient, inpatient hospital cost was $83,440 per patient, and ICU cost was $16,550 per patient. Discussion: In the US hospital setting, substantial HCRU and costs associated with intensive induction chemotherapy for AML were driven by inpatient hospitalizations.

Comparison of the 2021 COVID-19 roadmap projections against public health data in England

Control and mitigation of the COVID-19 pandemic in England has relied on a combination of vaccination and non-pharmaceutical interventions (NPIs). Some of these NPIs are extremely costly (economically and socially), so it was important to relax these promptly without overwhelming already burdened health services. The eventual policy was a Roadmap of four relaxation steps throughout 2021, taking England from lock-down to the cessation of all restrictions on social interaction. In a series of six Roadmap documents generated throughout 2021, models assessed the potential risk of each relaxation step. Here we show that the model projections generated a reliable estimation of medium-term hospital admission trends, with the data points up to September 2021 generally lying within our 95% prediction intervals. The greatest uncertainties in the modelled scenarios came from vaccine efficacy estimates against novel variants, and from assumptions about human behaviour in the face of changing restrictions and risk.

Lebenssituation und Versorgung von Menschen mit Diabetes mellitus: Ein Scoping Review zu den Auswirkungen der COVID-19-Pandemie in Deutschland

Die vorliegende Arbeit dient als Bestandsaufnahme der Auswirkungen der COVID-19-Pandemie auf die Lebens- und Versorgungssituation von Menschen mit Diabetes mellitus in Deutschland. Dazu wurde eine systematische Recherche nach Art eines Scoping Reviews durchgeführt. Zum einen wurde eine systematische Literaturrecherche in wissenschaftlichen Datenbanken nach empirischen Studien und mit anderen Suchinstrumenten nach nicht-empirischen Publikationen durchgeführt. Zum anderen wurden Routinedaten (z.B. GKV-Routinedaten, Daten aus Patientenregistern, vertragsärztliche Abrechnungs- und Arzneiversorgungsdaten) bei Krankenkassen, Patientenregistern oder anderen Institutionen angefragt, um Rückschlüsse auf die Versorgungssituation von Menschen mit Diabetes zu gewinnen.

Bei der Literaturrecherche wurden insgesamt 53 Veröffentlichungen (12 empirische Studien und 41 andere Publikationen) identifiziert und in die Datenextraktion eingeschlossen. Die empirischen Studien wurden zudem qualitativ bewertet. Aufgrund der geringen Anzahl empirischer Studien und ihrer niedrigen Qualität sind die Evidenzlücken zu den Auswirkungen der COVID-19-Pandemie auf die Versorgung von Menschen mit Diabetes groß. Allerdings liefern die empirischen Studien Anhaltspunkte dafür, dass sich die Pandemie auf die Inanspruchnahme von diabetesspezifischen Leistungen negativ ausgewirkt hat. Die Studien zeigen weniger Neu- und Wiedereinschreibungen in Disease-Management-Programme (DMP) für Diabetes; weniger Änderungen bei Verordnungen von blutglukosesenkenden Medikamenten; weniger Diabetes-Diagnosen und eine höhere Rate von diabetischen Ketoazidosen bei Kindern und Jugendlichen. Weiter zeigte sich, dass die COVID-19-Pandemie die Nutzung digitaler Möglichkeiten bei der Versorgung von Menschen mit Diabetes gefördert hat. Die Recherche nach Routinedaten blieb hingegen ergebnislos. Zusammenfassend fanden sich nur wenige belastbare Daten zu den Auswirkungen der COVID-19-Pandemie auf die Versorgung von Menschen mit Diabetes in Deutschland.

Associations of handgrip strength with morbidity and all-cause mortality of cardiometabolic multimorbidity

BACKGROUND

Cardiometabolic multimorbidity (CM) is an increasing public health and clinical concern. However, predictors for the development and prognosis of CM are poorly understood. The aims of this study were to investigate the relation between handgrip strength (HGS) and the risk of CM and to examine the association of HGS with all-cause mortality risk among patients with CM.

METHODS

This prospective cohort study involved 493,774 participants from the UK Biobank. CM was defined as the simultaneous occurrence of two or more of the following conditions: type 2 diabetes, stroke, and coronary heart disease (CHD). Cox proportional hazards models were performed to estimate hazard ratios (HRs) and 95% confidence intervals (95% CIs).

RESULTS

During a median follow-up of 12.1 years, 4701 incident CM cases were documented among participants with none cardiometabolic disease at baseline. Compared with the fourth quartile (Q4), the multivariable adjusted HR (95% CI) value of Q1 of HGS for developing CM was 1.46 (1.34-1.60). In participants with one cardiometabolic disease at baseline, participants in Q1 of HGS also possessed higher risk of CM than those in Q4, with HRs (95% CIs) being 1.35 (1.23-1.49) in patients with type 2 diabetes, 1.23 (1.04-1.46) in patients with stroke, and 1.23 (1.11-1.36) in patients with CHD. For participants with CM at recruitment, HGS was also associated with the risk of all-cause mortality (Q1 vs. Q4 HR: 1.57, 95% CI: 1.36-1.80).

CONCLUSIONS

Our study provided novel evidence that HGS could be an independent predictor of morbidity and all-cause mortality of CM.

A Comparison of Logistic Regression Against Machine Learning Algorithms for Gastric Cancer Risk Prediction Within Real-World Clinical Data Streams

PURPOSE

Noncardia gastric cancer (NCGC) is a leading cause of global cancer mortality, and is often diagnosed at advanced stages. Development of NCGC risk models within electronic health records (EHR) may allow for improved cancer prevention. There has been much recent interest in use of machine learning (ML) for cancer prediction, but few studies comparing ML with classical statistical models for NCGC risk prediction.

METHODS

We trained models using logistic regression (LR) and four commonly used ML algorithms to predict NCGC from age-/sex-matched controls in two EHR systems: Stanford University and the University of Washington (UW). The LR model contained well-established NCGC risk factors (intestinal metaplasia histology, prior Helicobacter pylori infection, race, ethnicity, nativity status, smoking history, anemia), whereas ML models agnostically selected variables from the EHR. Models were developed and internally validated in the Stanford data, and externally validated in the UW data. Hyperparameter tuning of models was achieved using cross-validation. Model performance was compared by accuracy, sensitivity, and specificity.

RESULTS

In internal validation, LR performed with comparable accuracy (0.732; 95% CI, 0.698 to 0.764), sensitivity (0.697; 95% CI, 0.647 to 0.744), and specificity (0.767; 95% CI, 0.720 to 0.809) to penalized lasso, support vector machine, K-nearest neighbor, and random forest models. In external validation, LR continued to demonstrate high accuracy, sensitivity, and specificity. Although K-nearest neighbor demonstrated higher accuracy and specificity, this was offset by significantly lower sensitivity. No ML model consistently outperformed LR across evaluation criteria.

CONCLUSION

Drawing data from two independent EHRs, we find LR on the basis of established risk factors demonstrated comparable performance to optimized ML algorithms. This study demonstrates that classical models built on robust, hand-chosen predictor variables may not be inferior to data-driven models for NCGC risk prediction.

Characterisation of type 2 diabetes subgroups and their association with ethnicity and clinical outcomes: a UK real-world data study using the East London Database

BACKGROUND

Subgroups of type 2 diabetes (T2DM) have been well characterised in experimental studies. It is unclear, however, whether the same approaches can be used to characterise T2DM subgroups in UK primary care populations and their associations with clinical outcomes.

AIM

To derive T2DM subgroups using primary care data from a multi-ethnic population, evaluate associations with glycaemic control, treatment initiation, and vascular outcomes, and to understand how these vary by ethnicity.

DESIGN AND SETTING

An observational cohort study in the East London Primary Care Database from 2008 to 2018.

METHOD

Latent-class analysis using age, sex, glycated haemoglobin, and body mass index at diagnosis was used to derive T2DM subgroups in white, South Asian, and black groups. Time to treatment initiation and vascular outcomes were estimated using multivariable Cox-proportional hazards regression.

RESULTS

In total, 31 931 adults with T2DM were included: 47% South Asian (n = 14 884), 26% white (n = 8154), 20% black (n = 6423). Two previously described subgroups were replicated, 'mild age-related diabetes' (MARD) and 'mild obesity-related diabetes' (MOD), and a third was characterised 'severe hyperglycaemic diabetes' (SHD). Compared with MARD, SHD had the poorest long-term glycaemic control, fastest initiation of antidiabetic treatment (hazard ratio [HR] 2.02, 95% confidence interval [CI] = 1.76 to 2.32), and highest risk of microvascular complications (HR 1.38, 95% CI = 1.28 to 1.49). MOD had the highest risk of macrovascular complications (HR 1.50, 95% CI = 1.23 to 1.82). Subgroup differences in treatment initiation were most pronounced for the white group, and vascular complications for the black group.

CONCLUSION

Clinically useful T2DM subgroups, identified at diagnosis, can be generated in routine real-world multi-ethnic populations, and may offer a pragmatic means to develop stratified primary care pathways and improve healthcare resource allocation.

Data mining to retrieve smoking status from electronic health records in general practice. EUROPEAN HEART JOURNAL. DIGITAL HEALTH 2022;3:437-444. [PMID: 36712169 PMCID: PMC9707867 DOI: 10.1093/ehjdh/ztac031]

Aims

Optimize and assess the performance of an existing data mining algorithm for smoking status from hospital electronic health records (EHRs) in general practice EHRs.

Methods and results

We optimized an existing algorithm in a training set containing all clinical notes from 498 individuals (75 712 contact moments) from the Julius General Practitioners' Network (JGPN). Each moment was classified as either 'current smoker', 'former smoker', 'never smoker', or 'no information'. As a reference, we manually reviewed EHRs. Algorithm performance was assessed in an independent test set (n = 494, 78 129 moments) using precision, recall, and F1-score. Test set algorithm performance for 'current smoker' was precision 79.7%, recall 78.3%, and F1-score 0.79. For former smoker, it was precision 73.8%, recall 64.0%, and F1-score 0.69. For never smoker, it was precision 92.0%, recall 74.9%, and F1-score 0.83. On a patient level, performance for ever smoker (current and former smoker combined) was precision 87.9%, recall 94.7%, and F1-score 0.91. For never smoker, it was 98.0, 82.0, and 0.89%, respectively. We found a more narrative writing style in general practice than in hospital EHRs.

Conclusion

Data mining can successfully retrieve smoking status information from general practice clinical notes with a good performance for classifying ever and never smokers. Differences between general practice and hospital EHRs call for optimization of data mining algorithms when applied beyond a primary development setting.

Use of Digital Health Interventions in Sub-Saharan Africa for Health Systems Strengthening Over the Last 10 Years: A Scoping Review Protocol

Background

Digital Health Interventions (DHIs) refers to the utilization of digital and mobile technology to support the health system in service delivery. Over the recent years, advanced computing, genomics, and artificial intelligence are considered part of digital health. In the context of the World Health Organization (WHO) global strategy 2020-2025, digital health is defined as "the field of knowledge and practice associated with the development and use of digital technologies to improve health." The scoping review protocol details the procedure for developing a comprehensive list of DHIs in Sub-Saharan Africa and documenting their roles in strengthening health systems.

Method and Analysis

A scoping review will be done according to the Joanne Briggs institute reviewers manual and following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) checklist and explanation. The protocol has been registered at the Open Science Framework (OSF) database at https://osf.io/5kzq7. The review will include DHIs conceptualized/developed/designed, adapted, piloted, deployed, scaled up, and addressing health challenges in Sub-Saharan Africa. We will retrieve data from the global DHI repository-the WHO Digital Health Atlas (DHA)- and supplement it with information from the WHO eHealth Observatory, eHealth Survey (2015), and eHealth country profiles report. Additional searches will be conducted in four (4) electronic databases: PubMed, HINARI-Reasearch4Life, Cochrane Library, and Google Scholar. The review will also include gray literature and reference lists of selected studies. Data will be organized in conceptual categories looking at digital health interventions' distinct function toward achieving health sector objectives.

Discussion

Sub-Saharan Africa is an emerging powerhouse in DHI innovations with rapid expansion and evolvement. The enthusiasm for digital health has experienced challenges including an escalation of short-lived digital health interventions, duplication, and minimal documentation of evidence on their impact on the health system. Efficient use of resources is important when striving toward the use digital health interventions in health systems strengthening. This can be achieved through documenting successes and lessons learnt over time.

Conclusion

The review will provide the evidence to guide further investments in DHIs, avoid duplication, circumvent barriers, focus on gaps, and scale-up successful interventions.

Implications for health system resilience: Quantifying the impact of the COVID-19-related stay at home orders on cancer screenings and diagnoses in southeastern North Carolina, USA

COVID-19 impacted hospital systems across the globe. Focus shifted to responding to increased healthcare demand while mitigating COVID-19 spread on their campuses. Mitigation efforts limited medical professional-patient interactions, including patient access to preventive cancer screenings. Data were gleaned from a health information exchange containing records on over 2 million patients in southeastern North Carolina, USA. This study tested five hypotheses: H₁: Weekly cancer screenings significantly decreased during North Carolina's (NC) Stay-At-Home (SAH) orders; H₂: Weekly cancer diagnoses significantly decreased during NC's SAH orders; H₃: Weekly cancer screenings significantly increased after the end of NC's SAH orders; H₄: Weekly cancer diagnoses significantly increased after the end of NC's SAH orders; and H₅: Weekly advanced cancer diagnoses significantly increased after the end of NC's SAH orders. Time series regression analysis was employed to quantify trends. Results suggested strong support of H₁ and H₃, moderate support of H₄, mixed support of H₅, and no support of H₂. For example, compared to before the SAH orders, we estimated 662.3 fewer weekly breast cancer screenings during the SAH orders (H₁). After the SAH orders (H₃), we estimated 232.5 more breast cancer screenings and 10.6 more breast cancer diagnoses. This work quantifies the impact of COVID-19 associated SAH orders on cancer screenings and diagnoses and suggests the potential for delayed or missed cancer diagnoses. This evident disruption in providing routine medical care also highlights the importance of strengthening health systems (or organizations) and improving resilience to natural disasters and infectious disease outbreaks.

Immortal time bias for life-long conditions in retrospective observational studies using electronic health records

Background

Immortal time bias is common in observational studies but is typically described for pharmacoepidemiology studies where there is a delay between cohort entry and treatment initiation.

Methods

This study used the Clinical Practice Research Datalink (CPRD) and linked national mortality data in England from 2000 to 2019 to investigate immortal time bias for a specific life-long condition, intellectual disability. Life expectancy (Chiang’s abridged life table approach) was compared for 33,867 exposed and 980,586 unexposed individuals aged 10+ years using five methods: (1) treating immortal time as observation time; (2) excluding time before date of first exposure diagnosis; (3) matching cohort entry to first exposure diagnosis; (4) excluding time before proxy date of inputting first exposure diagnosis (by the physician); and (5) treating exposure as a time-dependent measure.

Results

When not considered in the design or analysis (Method 1), immortal time bias led to disproportionately high life expectancy for the exposed population during the first calendar period (additional years expected to live: 2000–2004: 65.6 [95% CI: 63.6,67.6]) compared to the later calendar periods (2005–2009: 59.9 [58.8,60.9]; 2010–2014: 58.0 [57.1,58.9]; 2015–2019: 58.2 [56.8,59.7]). Date of entry of diagnosis (Method 4) was unreliable in this CPRD cohort. The final methods (Method 2, 3 and 5) appeared to solve the main theoretical problem but residual bias may have remained.

Conclusions

We conclude that immortal time bias is a significant issue for studies of life-long conditions that use electronic health record data and requires careful consideration of how clinical diagnoses are entered onto electronic health record systems.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12874-022-01581-1.

Using data science to improve outcomes for persons with opioid use disorder

Medication treatment for opioid use disorder (MOUD) is an effective evidence-based therapy for decreasing opioid-related adverse outcomes. Effective strategies for retaining persons on MOUD, an essential step to improving outcomes, are needed as roughly half of all persons initiating MOUD discontinue within a year. Data science may be valuable and promising for improving MOUD retention by using "big data" (e.g., electronic health record data, claims data mobile/sensor data, social media data) and specific machine learning techniques (e.g., predictive modeling, natural language processing, reinforcement learning) to individualize patient care. Maximizing the utility of data science to improve MOUD retention requires a three-pronged approach: (1) increasing funding for data science research for OUD, (2) integrating data from multiple sources including treatment for OUD and general medical care as well as data not specific to medical care (e.g., mobile, sensor, and social media data), and (3) applying multiple data science approaches with integrated big data to provide insights and optimize advances in the OUD and overall addiction fields.

Validation of algorithms for identifying outpatient infections in MS patients using electronic medical records

Background Our multiple sclerosis (MS) stakeholder groups expressed concerns about whether MS disease-modifying therapies (DMTs) increase the risk of specific outpatient infections. Validated methods for identifying the risk of these selected outpatient infections in the general population either do not exist, exclude the clinically important possibility of recurrent infections, or are inaccurate, largely because existing studies relied primarily on International Classification of Diseases (ICD) codes to identify infectious outcomes. Additionally, no studies have validated methods among the MS population, where some MS symptoms can be mistaken for infections (e.g., urinary tract infections (UTIs)). Objective To utilize multiple data elements in the electronic health record (EHR) to improve accurate identification of selected outpatient infections in an MS cohort and general population controls. Methods We searched Kaiser Permanente Southern California's EHR based on ICD-9/10 codes for specified outpatient infections from 1/1/2008-12/31/2018 among our MS cohort (n=6000) and 5:1 general population controls matched on age, sex, and race/ethnicity (n=30,010). Random sample chart abstractions from each group were used to identify common coding errors for outpatient pneumonia, upper and lower respiratory tract infection, UTIs, herpetic infections (herpes zoster (HZ), herpes simplex virus (HSV)), fungal infections, otitis media, cellulitis, and influenza. This information was used to define discrete infectious episodes and to identify the algorithm with the highest positive predictive value (PPV) after supplementing the ICD-coded episodes with radiology, laboratory and/or pharmacy data. Results PPVs relying on ICD codes alone were inaccurate, particularly for identifying recurrent herpetic infections (HZ (42%) and HSV (60%)), UTIs (42%) and outpatient pneumonia (20%) in MS patients. Defining and validating episodes improved the PPVs for all the selected infections. The final algorithms' PPVs were 80-100% in MS and 75-100% in the general population, after including dispensed treatments (UTI, herpetic infections and yeast vaginitis), timing of dispensed treatments (UTI, herpetic infections and yeast vaginitis), removal of prophylactic antiviral use (herpetic infections), and inclusion of selected laboratory (UTIs) and imaging results (pneumonia). The only exception was outpatient pneumonia, where PPVs improved but remained ≤70%. There were no significant differences in the PPVs for the final algorithms between the MS and general population. Conclusions Provided herein are accurate and validated algorithms that can be used to improve our understanding of how the risk of recurrent outpatient infections are influenced by MS treatments, MS-related disability, and co-morbidities. Findings from such studies will be important in helping patients and clinicians engage in shared decision-making and in developing strategies to mitigate risks of recurrent infections.

How can we discover the most valuable types of big data and artificial intelligence-based solutions? A methodology for the efficient development of the underlying analytics that improve care

BACKGROUND

Much has been invested in big data and artificial intelligence-based solutions for healthcare. However, few applications have been implemented in clinical practice. Early economic evaluations can help to improve decision-making by developers of analytics underlying these solutions aiming to increase the likelihood of successful implementation, but recommendations about their use are lacking. The aim of this study was to develop and apply a framework that positions best practice methods for economic evaluations alongside development of analytics, thereby enabling developers to identify barriers to success and to select analytics worth further investments.

METHODS

The framework was developed using literature, recommendations for economic evaluations and by applying the framework to use cases (chronic lymphocytic leukaemia (CLL), intensive care, diabetes). First, the feasibility of developing clinically relevant analytics was assessed and critical barriers to successful development and implementation identified. Economic evaluations were then used to determine critical thresholds and guide investment decisions.

RESULTS

When using the framework to assist decision-making of developers of analytics, continuing development was not always feasible or worthwhile. Developing analytics for progressive CLL and diabetes was clinically relevant but not feasible with the data available. Alternatively, developing analytics for newly diagnosed CLL patients was feasible but continuing development was not considered worthwhile because the high drug costs made it economically unattractive for potential users. Alternatively, in the intensive care unit, analytics reduced mortality and per-patient costs when used to identify infections (- 0.5%, - €886) and to improve patient-ventilator interaction (- 3%, - €264). Both analytics have the potential to save money but the potential benefits of analytics that identify infections strongly depend on infection rate; a higher rate implies greater cost-savings.

CONCLUSIONS

We present a framework that stimulates efficiency of development of analytics for big data and artificial intelligence-based solutions by selecting those applications of analytics for which development is feasible and worthwhile. For these applications, results from early economic evaluations can be used to guide investment decisions and identify critical requirements.

Effect of COVID-19 pandemic lockdowns on planned cancer surgery for 15 tumour types in 61 countries: an international, prospective, cohort study

BACKGROUND

Surgery is the main modality of cure for solid cancers and was prioritised to continue during COVID-19 outbreaks. This study aimed to identify immediate areas for system strengthening by comparing the delivery of elective cancer surgery during the COVID-19 pandemic in periods of lockdown versus light restriction.

METHODS

This international, prospective, cohort study enrolled 20 006 adult (≥18 years) patients from 466 hospitals in 61 countries with 15 cancer types, who had a decision for curative surgery during the COVID-19 pandemic and were followed up until the point of surgery or cessation of follow-up (Aug 31, 2020). Average national Oxford COVID-19 Stringency Index scores were calculated to define the government response to COVID-19 for each patient for the period they awaited surgery, and classified into light restrictions (index <20), moderate lockdowns (20-60), and full lockdowns (>60). The primary outcome was the non-operation rate (defined as the proportion of patients who did not undergo planned surgery). Cox proportional-hazards regression models were used to explore the associations between lockdowns and non-operation. Intervals from diagnosis to surgery were compared across COVID-19 government response index groups. This study was registered at ClinicalTrials.gov, NCT04384926.

FINDINGS

Of eligible patients awaiting surgery, 2003 (10·0%) of 20 006 did not receive surgery after a median follow-up of 23 weeks (IQR 16-30), all of whom had a COVID-19-related reason given for non-operation. Light restrictions were associated with a 0·6% non-operation rate (26 of 4521), moderate lockdowns with a 5·5% rate (201 of 3646; adjusted hazard ratio [HR] 0·81, 95% CI 0·77-0·84; p<0·0001), and full lockdowns with a 15·0% rate (1775 of 11 827; HR 0·51, 0·50-0·53; p<0·0001). In sensitivity analyses, including adjustment for SARS-CoV-2 case notification rates, moderate lockdowns (HR 0·84, 95% CI 0·80-0·88; p<0·001), and full lockdowns (0·57, 0·54-0·60; p<0·001), remained independently associated with non-operation. Surgery beyond 12 weeks from diagnosis in patients without neoadjuvant therapy increased during lockdowns (374 [9·1%] of 4521 in light restrictions, 317 [10·4%] of 3646 in moderate lockdowns, 2001 [23·8%] of 11 827 in full lockdowns), although there were no differences in resectability rates observed with longer delays.

INTERPRETATION

Cancer surgery systems worldwide were fragile to lockdowns, with one in seven patients who were in regions with full lockdowns not undergoing planned surgery and experiencing longer preoperative delays. Although short-term oncological outcomes were not compromised in those selected for surgery, delays and non-operations might lead to long-term reductions in survival. During current and future periods of societal restriction, the resilience of elective surgery systems requires strengthening, which might include protected elective surgical pathways and long-term investment in surge capacity for acute care during public health emergencies to protect elective staff and services.

FUNDING

National Institute for Health Research Global Health Research Unit, Association of Coloproctology of Great Britain and Ireland, Bowel and Cancer Research, Bowel Disease Research Foundation, Association of Upper Gastrointestinal Surgeons, British Association of Surgical Oncology, British Gynaecological Cancer Society, European Society of Coloproctology, Medtronic, Sarcoma UK, The Urology Foundation, Vascular Society for Great Britain and Ireland, and Yorkshire Cancer Research.

A narrative review on the validity of electronic health record-based research in epidemiology

Electronic health records (EHRs) are widely used in epidemiological research, but the validity of the results is dependent upon the assumptions made about the healthcare system, the patient, and the provider. In this review, we identify four overarching challenges in using EHR-based data for epidemiological analysis, with a particular emphasis on threats to validity. These challenges include representativeness of the EHR to a target population, the availability and interpretability of clinical and non-clinical data, and missing data at both the variable and observation levels. Each challenge reveals layers of assumptions that the epidemiologist is required to make, from the point of patient entry into the healthcare system, to the provider documenting the results of the clinical exam and follow-up of the patient longitudinally; all with the potential to bias the results of analysis of these data. Understanding the extent of as well as remediating potential biases requires a variety of methodological approaches, from traditional sensitivity analyses and validation studies, to newer techniques such as natural language processing. Beyond methods to address these challenges, it will remain crucial for epidemiologists to engage with clinicians and informaticians at their institutions to ensure data quality and accessibility by forming multidisciplinary teams around specific research projects.

Performance of a Machine Learning Algorithm Using Electronic Health Record Data to Identify and Estimate Survival in a Longitudinal Cohort of Patients With Lung Cancer

IMPORTANCE

Electronic health records (EHRs) provide a low-cost means of accessing detailed longitudinal clinical data for large populations. A lung cancer cohort assembled from EHR data would be a powerful platform for clinical outcome studies.

OBJECTIVE

To investigate whether a clinical cohort assembled from EHRs could be used in a lung cancer prognosis study.

DESIGN, SETTING, AND PARTICIPANTS

In this cohort study, patients with lung cancer were identified among 76 643 patients with at least 1 lung cancer diagnostic code deposited in an EHR in Mass General Brigham health care system from July 1988 to October 2018. Patients were identified via a semisupervised machine learning algorithm, for which clinical information was extracted from structured and unstructured data via natural language processing tools. Data completeness and accuracy were assessed by comparing with the Boston Lung Cancer Study and against criterion standard EHR review results. A prognostic model for non-small cell lung cancer (NSCLC) overall survival was further developed for clinical application. Data were analyzed from March 2019 through July 2020.

EXPOSURES

Clinical data deposited in EHRs for cohort construction and variables of interest for the prognostic model were collected.

MAIN OUTCOMES AND MEASURES

The primary outcomes were the performance of the lung cancer classification model and the quality of the extracted variables; the secondary outcome was the performance of the prognostic model.

RESULTS

Among 76 643 patients with at least 1 lung cancer diagnostic code, 42 069 patients were identified as having lung cancer, with a positive predictive value of 94.4%. The study cohort consisted of 35 375 patients (16 613 men [47.0%] and 18 756 women [53.0%]; 30 140 White individuals [85.2%], 1040 Black individuals [2.9%], and 857 Asian individuals [2.4%]) after excluding patients with lung cancer history and less than 14 days of follow-up after initial diagnosis. The median (interquartile range) age at diagnosis was 66.7 (58.4-74.1) years. The area under the receiver operating characteristic curves of the prognostic model for overall survival with NSCLC were 0.828 (95% CI, 0.815-0.842) for 1-year prediction, 0.825 (95% CI, 0.812-0.836) for 2-year prediction, 0.814 (95% CI, 0.800-0.826) for 3-year prediction, 0.814 (95% CI, 0.799-0.828) for 4-year prediction, and 0.812 (95% CI, 0.798-0.825) for 5-year prediction.

CONCLUSIONS AND RELEVANCE

These findings suggest the feasibility of assembling a large-scale EHR-based lung cancer cohort with detailed longitudinal clinical measurements and that EHR data may be applied in cancer progression with a set of generalizable approaches.

Ethnic differences in guideline-indicated statin initiation for people with type 2 diabetes in UK primary care, 2006-2019: A cohort study

BACKGROUND

Type 2 diabetes is 2-3 times more prevalent in people of South Asian and African/African Caribbean ethnicity than people of European ethnicity living in the UK. The former 2 groups also experience excess atherosclerotic cardiovascular disease (ASCVD) complications of diabetes. We aimed to study ethnic differences in statin initiation, a cornerstone of ASCVD primary prevention, for people with type 2 diabetes.

METHODS AND FINDINGS

Observational cohort study of UK primary care records, from 1 January 2006 to 30 June 2019. Data were studied from 27,511 (88%) people of European ethnicity, 2,386 (8%) people of South Asian ethnicity, and 1,142 (4%) people of African/African Caribbean ethnicity with incident type 2 diabetes, no previous ASCVD, and statin use indicated by guidelines. Statin initiation rates were contrasted by ethnicity, and the number of ASCVD events that could be prevented by equalising prescribing rates across ethnic groups was estimated. Median time to statin initiation was 79, 109, and 84 days for people of European, South Asian, and African/African Caribbean ethnicity, respectively. People of African/African Caribbean ethnicity were a third less likely to receive guideline-indicated statins than European people (n/N [%]: 605/1,142 [53%] and 18,803/27,511 [68%], respectively; age- and gender-adjusted HR 0.67 [95% CI 0.60 to 0.76], p < 0.001). The HR attenuated marginally in a model adjusting for total cholesterol/high-density lipoprotein cholesterol ratio (0.77 [95% CI 0.69 to 0.85], p < 0.001), with no further diminution when deprivation, ASCVD risk factors, comorbidity, polypharmacy, and healthcare usage were accounted for (fully adjusted HR 0.76 [95% CI 0.68, 0.85], p < 0.001). People of South Asian ethnicity were 10% less likely to receive a statin than European people (1,489/2,386 [62%] and 18,803/27,511 [68%], respectively; fully adjusted HR 0.91 [95% CI 0.85 to 0.98], p = 0.008, adjusting for all covariates). We estimated that up to 12,600 ASCVD events could be prevented over the lifetimes of people currently affected by type 2 diabetes in the UK by equalising statin prescribing across ethnic groups. Limitations included incompleteness of recording of routinely collected data.

CONCLUSIONS

In this study we observed that people of African/African Caribbean ethnicity with type 2 diabetes were substantially less likely, and people of South Asian ethnicity marginally less likely, to receive guideline-indicated statins than people of European ethnicity, even after accounting for sociodemographics, healthcare usage, ASCVD risk factors, and comorbidity. Underuse of statins in people of African/African Caribbean or South Asian ethnicity with type 2 diabetes is a missed opportunity to prevent cardiovascular events.

Leveraging unstructured data to identify hereditary angioedema patients in electronic medical records

Background

The epidemiologic impact of hereditary angioedema (HAE) is difficult to quantify, due to misclassification in retrospective studies resulting from non-specific diagnostic coding. The aim of this study was to identify cohorts of patients with HAE-1/2 by evaluating structured and unstructured data in a US ambulatory electronic medical record (EMR) database.

Methods

A retrospective feasibility study was performed using the GE Centricity EMR Database (2006–2017). Patients with ≥ 1 diagnosis code for HAE-1/2 (International Classification of Diseases, Ninth Revision, Clinical Modification 277.6 or International Classification of Diseases, Tenth Revision, Clinical Modification D84.1) and/or ≥ 1 physician note regarding HAE-1/2 and ≥ 6 months’ data before and after the earliest code or note (index date) were included. Two mutually exclusive cohorts were created: probable HAE (≥ 2 codes or ≥ 2 notes on separate days) and suspected HAE (only 1 code or note). The impact of manually reviewing physician notes on cohort formation was assessed, and demographic and clinical characteristics of the 2 final cohorts were described.

Results

Initially, 1691 patients were identified: 190 and 1501 in the probable and suspected HAE cohorts, respectively. After physician note review, the confirmed HAE cohort comprised 254 patients and the suspected HAE cohort decreased to 1299 patients; 138 patients were determined not to have HAE and were excluded. The overall false-positive rate for the initial algorithms was 8.2%. Across final cohorts, the median age was 50 years and > 60% of patients were female. HAE-specific prescriptions were identified for 31% and 2% of the confirmed and suspected HAE cohorts, respectively.

Conclusions

Unstructured EMR data can provide valuable information for identifying patients with HAE-1/2. Further research is needed to develop algorithms for more representative HAE cohorts in retrospective studies.

Measuring and Quantifying Collateral Information in Psychiatry: Development and Preliminary Validation of the McLean Collateral Information and Clinical Actionability Scale

BACKGROUND

The review of collateral information is an essential component of patient care. Although this is standard practice, minimal research has been done to quantify collateral information collection and to understand how collateral information translates to clinical decision making. To address this, we developed and piloted a novel measure (the McLean Collateral Information and Clinical Actionability Scale [M-CICAS]) to evaluate the types and number of collateral sources viewed and the resulting actions made in a psychiatric setting.

OBJECTIVE

This study aims to test the feasibility of the M-CICAS, validate this measure against clinician notes via medical records, and evaluate whether reviewing a higher volume of collateral sources is associated with more clinical actions taken.

METHODS

For the M-CICAS, we developed a three-part instrument, focusing on measuring collateral sources reviewed, clinical actions taken, and shared decision making between the clinician and patient. To determine feasibility and preliminary validity, we piloted this measure among clinicians providing psychotherapy at McLean Hospital. These clinicians (n=7) completed the M-CICAS after individual clinical sessions with 89 distinct patient encounters. Scales were completed by clinicians only once for each patient during routine follow-up visits. After clinicians completed these scales, researchers conducted chart reviews by completing the M-CICAS using only the clinician's corresponding note from that session. For the analyses, we generated summary scores for the number of collateral sources and clinical actions for each encounter. We examined Pearson correlation coefficients to assess interrater reliability between clinicians and chart reviewers, and simple univariate regression modeling followed by multilevel mixed effects regression modeling to test the relationship between collateral information accessed and clinical actions taken.

RESULTS

The study staff had high interrater reliability on the M-CICAS for the sources reviewed (r=0.98; P<.001) and actions taken (r=0.97; P<.001). Clinician and study staff ratings were moderately correlated and statistically significant on the M-CICAS summary scores for the sources viewed (r=0.24, P=.02 and r=0.25, P=.02, respectively). Univariate regression modeling with a two-tailed test demonstrated a significant association between collateral sources and clinical actions taken when clinicians completed the M-CICAS (β=.27; t₈₇=2.47; P=.02). The multilevel fixed slopes random intercepts model confirmed a significant association even when accounting for clinician differences (β=.23; t₅₇=2.13; P=.04).

CONCLUSIONS

This pilot study established the feasibility and preliminary validity of the M-CICAS in assessing collateral sources and clinical decision making in psychiatry. This study also indicated that reviewing more collateral sources may lead to an increased number of clinical actions following a session.