Diagnostic and Translational Utility of the Secondary Traumatic Stress Clinical Algorithm (STS-CA)

Current tools available to assess secondary traumatic stress (STS) do not account for whether the symptoms are functionally related to indirect trauma, determine functional impairment caused by the STS symptoms, and/or consider the duration of the disturbance. This prevents delineation of various expressions of traumatic stress related to indirect trauma that may constitute the phenomenon of STS. The STS Clinical Algorithm (STS-CA) was developed to make these distinctions, so that interventions can be tailored to need. This study investigates the following: (1) the diagnostic concordance between the STS-CA findings and scores on the Secondary Traumatic Stress Scale (STSS); (2) reasons for diagnostic discrepancies between the STS-CA and the STSS assessments. Three trained interviewers used the STS-CA to guide the determination of clinical outcome (N = 181) in a diverse group of helping professionals. There was 100% agreement between the CAPS and the STS-CA, and fair agreement (κ =.426, p = .000) between the STS-CA and the STSS. The STS-CA demonstrated more sensitivity in classifying positive cases, and specificity in delineating those with atypical cluster presentations or little to no functional impairment that prohibited a post-traumatic stress disorder diagnosis than the STSS. Effective treatment of STS requires proper identification and the delivery of protocols that are tailored to the unique ways that STS manifests. This study provides some insights into the utility of the STS-CA in guiding this process and creates STS categories to organize and classify intervention strategies.

Impact of intracoronary assessments on revascularization decisions: A contemporary evaluation

OBJECTIVES

To investigate the real-world implementation of intracoronary assessment (ICA) techniques and evaluate their impact on clinical decisions regarding the management of coronary artery disease (CAD) in contemporary practice.

BACKGROUND

Coronary angiogram is the gold standard used to diagnose vessel stenosis and guide percutaneous coronary intervention (PCI); however, it is limited by its two-dimensional imaging capabilities. ICA techniques like intravascular ultrasound and optical coherence tomography capture the vessel in three-dimensional images. Comparatively, fractional flow reserve provides information on the physiologic significance of coronary stenosis. Both techniques may improve PCI outcomes if they routinely change physician behavior.

METHODS

Patients who underwent ICA between August 2015 and March 2020 were included in the study. The primary outcome was the clinical impact of ICA on physician clinical decision making of a stenotic vessel. The secondary outcome was the clinical changes that occurred following ICA.

RESULTS

A total of 1135 patients were included in the study. Physiologic assessment (PA) and image assessment (IA) were performed in 61.4% and 38.6% respectively. Management plans were changed in 38.1% and 23.9% of patients who received PA and IA. Over half of the management change resulted in physicians deciding to not intervene on the stenotic vessel. One-year outcome of these decisions showed no significant increase in major adverse cardiac events (hazard ratio [HR], 0.68; 95% confidence interval [CI], 0.40-1.15; p = 0.15) or unplanned revascularization (HR, 0.78; 95% CI, 0.35-1.74; p = 0.55) suggesting reliance on PA/IA data did not increase risk.

CONCLUSION

Selected ICA alters physician management of CAD in one-third of patients being evaluated for revascularization-typically leading to fewer interventions. All cause death is numerally lower in patients that received a change in management. However, the 1-year outcome of these altered decisions does not appear to be significantly different.

Development and validation of a nomogram for esophagogastric variceal bleeding in liver cirrhosis: A cohort study in 1099 cases

OBJECTIVES

Esophagogastric variceal bleeding (EVB) is a catastrophic complication of decompensated liver cirrhosis. We aimed to establish a nomogram based on noninvasive clinical and imaging variables to predict the risk of EVB.

METHODS

The cut-off value of each variable was determined through univariate regression analysis. The least absolute shrinkage and selection operator (LASSO) regression and multivariate logistic regression analyses were used to determine the risk factors and establish predictive models. The nomogram was established and validated using the calibration discrimination across different groups.

RESULTS

Six indicators, including platelet count, hemoglobin, albumin to globulin ratio, fasting blood glucose, serum chloride, and computed tomography portal vein diameter (CTPD), were found to be related to the risk of EVB. Two models, with or without CTPD, were established and compared. Model 1 with CTPD had better discrimination than model 2 with C-index of 0.893 (95% confidence interval [CI] 0.872-0.915) and 0.862 (95% CI 0.837-0.887) in the primary cohort, respectively (Z = 2.027, P = 0.043). While the C-index of the two models in the validation cohort was 0.878 (95% CI 0.838-0.919) and 0.810 (95% CI 0.757-0.863). Moreover, the clinical decision analysis curve and clinical impact curve showed that these models might confer a significant net benefit on patients and provide a reference threshold for clinicians.

CONCLUSION

A nomogram using routine clinical indicators was established to predict the risk of EVB in patients with liver cirrhosis, which was verified in an independent cohort and demonstrated a great consistency.

Nuclear Medicine and Artificial Intelligence: Best Practices for Evaluation (the RELAINCE Guidelines)

An important need exists for strategies to perform rigorous objective clinical-task-based evaluation of artificial intelligence (AI) algorithms for nuclear medicine. To address this need, we propose a 4-class framework to evaluate AI algorithms for promise, technical task-specific efficacy, clinical decision making, and postdeployment efficacy. We provide best practices to evaluate AI algorithms for each of these classes. Each class of evaluation yields a claim that provides a descriptive performance of the AI algorithm. Key best practices are tabulated as the RELAINCE (Recommendations for EvaLuation of AI for NuClear medicinE) guidelines. The report was prepared by the Society of Nuclear Medicine and Molecular Imaging AI Task Force Evaluation team, which consisted of nuclear-medicine physicians, physicists, computational imaging scientists, and representatives from industry and regulatory agencies.

Transforming and extending library services by embracing technology and collaborations: A case study

Technology advances and collaborations with information technology and computer science groups have enabled library services to expand into new domains. Listening to user needs, eliminating administrative burden and saving users time remain strong foundations on which to build new library services enabled by technology. Examples of what is now possible is described, including service to user groups, successes, failures and challenges. Although technology advances have enabled library service enhancements to all user groups, special emphasis on new library services in support of the research enterprise is discussed. As Lindberg and Humphreys predicted in 2015, the research enterprise's need for responsible curation of research data has created new opportunities for library services and examples of those services are discussed. As technology continues to advance, new library services are expected to emerge. These may include regulatory and compliance services. By developing these services with user feedback to save users time and expedite their work, and in collaboration with technology experts, libraries can expect to offer sustainable and valued services for years to come.

Does previous prostate surgery affect multiparametric magnetic resonance imaging accuracy in detecting clinically significant prostate cancer? Results from a single institution series

BACKGROUND

Multiparametric MRI (mpMRI) has demonstrated high diagnostic accuracy for clinically significant PCa (csPCa). However, the accuracy of this test in men that received a previous prostatic surgery is still controversial. We aimed at assessing the effect of previous prostatic surgery on the detection of csPCa in a tertiary referral center.

METHOD

We relied on a cohort of 311 men with a positive mpMRI (prostate imaging - reporting and data system [PI-RADS] ≥ 3) who underwent a targeted (TBx) plus concomitant systematic random biopsy (SBx) at a single tertiary referral center between 2017 and 2020. The study outcome was to compare the detection of csPCa (Gleason score ≥ 3 + 4) between the two groups (no previous prostate surgery [Group 1] vs. previous prostate surgery [Group 2]). Multivariable logistic regression analysis (MVA) was used to assess the relationship between previous prostate surgery and the detection of csPCa at TBx, after taking into account potential clinical confounders.

RESULTS

Overall, 24 (8%) patients received a previous prostate surgery before undergoing mpMRI. Median prostate-specific antigen density was 0.15 versus 0.08 ng/ml/cc, in Group 1 versus 2, respectively. The most frequent finding at mpMRI was in Group 1 versus 2, PI-RADS 4 (55%) versus PI-RADS 3 and 4 (42% each). The majority of patients were biopsy naïve in both Groups 1 (66%) and 2 (71%). The overall detection of csPCa in Group 1 versus 2 was 83% versus 75%, respectively. Differently, the detection of csPCa at TBx in Groups 1 versus 2 was 76% versus 71%, respectively. At MVA, previous prostate surgery (odds ratio: 0.65; p = 0.02) was significantly associated with lower csPCa detection at TBx, after accounting for potential confounders.

CONCLUSION

The presence of previous prostate surgery significantly decreases the accuracy of mpMRI in detecting csPCa. These results should be taken into account when assessing patients with a history of prostatic surgery and a suspicious lesion at mpMRI, to better select those who might avoid an unnecessary biopsy.

MMF/MPA Is the Main Mediator of a Delayed Humoral Response With Reduced Antibody Decline in Kidney Transplant Recipients After SARS-CoV-2 mRNA Vaccination

Kidney transplant recipients (KTR) show significantly lower seroconversion rates after SARS-CoV-2 mRNA vaccination compared to dialysis patients (DP). Mycophenolate mofetil or mycophenolic acid (MMF/MPA) in particular has been identified as a risk factor for seroconversion failure. While the majority of all KTR worldwide receive MMF/MPA for immunosuppressive therapy, its impact on antibody decline in seroconverted KTR still remains unclear. In an observational study (NCT04799808), we investigated whether 132 seroconverted KTR (anti-spike S1 IgG or IgA positive after 2 vaccinations) show a more rapid antibody decline with MMF/MPA than those without this medication. A total of 2 months after mRNA vaccination, average anti-spike S1 IgG levels of KTR with MMF/MPA were lower than without (p = 0.001), while no differences between these two groups were observed after 6 months (p = 0.366). Similar results were obtained for anti-RBD IgG antibodies (T2 p = 0.003 and T3 p = 0.135). The probability of severe IgG decline with MMF/MPA was three times lower than without (p = 0.003, OR 0.236, 95% CI 0.091-0.609). In the multivariate analysis, neither immunosuppressants, such as calcineurin inhibitors, mTOR inhibitors (mTOR-I; mechanistic target of rapamycin), glucocorticoids, nor vaccine type, sex, or age showed a significant influence on IgG titer decline between 2 and 6 months. For the decision on additional booster vaccinations, we consider immunosurveillance to be needed as an integral part of renal transplant follow-up after SARS-CoV-2 mRNA vaccination. Not only the lack of seroconversion but also the peak and titer decline of the specific IgG and RBD IgG antibody formation after two mRNA vaccinations is significantly influenced by MMF/MPA.

The Utility of Myotonometry in Musculoskeletal Rehabilitation and Human Performance Programming

Myotonometry is a relatively novel method used to quantify the biomechanical and viscoelastic properties (stiffness, compliance, tone, elasticity, creep, mechanical relaxation) of palpable musculotendinous structures with portable mechanical devices called myotonometers. Myotonometers obtain these measures by recording the magnitude of radial tissue deformation that occurs in response to the amount of force that is perpendicularly applied to the tissue through a device's probe. Myotonometric parameters such as stiffness and compliance have repeatedly demonstrated strong correlations with force production and muscle activation. Paradoxically, individual muscle stiffness measures have been associated with both superior athletic performance and higher incidence of injury. This suggest there may be optimal stiffness levels that promotes athletic performance while too much or too little may lead to an increased risk of injury. Numerous studies suggest that myotonometry may assist practitioners in the development of performance and rehabilitation programs that improves athletic performance, mitigates injury risk, guides therapeutic interventions, and optimizes return to activity decision making. Thus, the purpose of this narrative review is to summarize the potential utility of myotonometry as a clinical tool that assists musculoskeletal clinicians with the diagnosis, rehabilitation, and prevention of athletic injuries.

Trials in the Key of G: Building Level 1 Evidence on the Real-world Effectiveness of Prostate Biomarkers

A number of genomic classifiers are available to aid in shared decision-making for men with localized prostate cancer; however, there is no high-level evidence assessing their clinical utility. The two randomized controlled trials in this report prospectively evaluate the use of gene expression classifier testing at the time of cancer diagnosis and after surgical treatment.

Provider Perspectives on Factors Affecting the PrEP Care Continuum Among Black Cisgender Women in the Midwest United States: Applying the Consolidated Framework for Implementation Research

BACKGROUND

To End the HIV Epidemic and reduce the number of incident HIV infections in the United States by 90%, pre-exposure prophylaxis (PrEP) uptake and persistence among cisgender women, particularly racial and ethnic minority women, must be increased. Medical providers play a pivotal role across the PrEP care continuum.

METHODS

In this qualitative study, guided by the Consolidated Framework for Implementation Research, we explored health care provider perspectives on facilitators and barriers to PrEP implementation strategies for Black cisgender women in the Midwest United States. Data were analyzed using a deductive thematic content analysis approach.

RESULTS

A total of 10 medical providers completed individual qualitative interviews. Using the Consolidated Framework for Implementation Research framework, we identified intervention characteristics (cost, dosing, and adherence), individual patient and provider level factors (self-efficacy, knowledge, and attitudes), and systematic barriers (inner setting and outer setting) that ultimately lead to PrEP inequalities. Implementation strategies to improve the PrEP care continuum identified include provider training, electronic medical record optimization, routine patient education, and PrEP navigation.

CONCLUSION

This study provides (1) medical provider insight into implementation factors that can be modified to improve the PrEP care continuum for Black cisgender women and (2) an implementation research logic model to guide future studies.

Machine Learning-Based Risk Stratification for Gestational Diabetes Management

Gestational diabetes mellitus (GDM) is often diagnosed during the last trimester of pregnancy, leaving only a short timeframe for intervention. However, appropriate assessment, management, and treatment have been shown to reduce the complications of GDM. This study introduces a machine learning-based stratification system for identifying patients at risk of exhibiting high blood glucose levels, based on daily blood glucose measurements and electronic health record (EHR) data from GDM patients. We internally trained and validated our model on a cohort of 1148 pregnancies at Oxford University Hospitals NHS Foundation Trust (OUH), and performed external validation on 709 patients from Royal Berkshire Hospital NHS Foundation Trust (RBH). We trained linear and non-linear tree-based regression models to predict the proportion of high-readings (readings above the UK's National Institute for Health and Care Excellence [NICE] guideline) a patient may exhibit in upcoming days, and found that XGBoost achieved the highest performance during internal validation (0.021 [CI 0.019-0.023], 0.482 [0.442-0.516], and 0.112 [0.109-0.116], for MSE, R2, MAE, respectively). The model also performed similarly during external validation, suggesting that our method is generalizable across different cohorts of GDM patients.

Managing IBD in patients with previous cancers

A frequent dilemma faced in the inflammatory bowel disease (IBD) clinic is how to best treat a patient with a previous cancer diagnosis. The changing demographics of our patient population will make this quandary more common. Previous guidance has emphasised the importance of lengthy postcancer drug holidays and cautious use of IBD therapies. However, accumulating evidence suggests this approach may be unnecessarily conservative. This review considers recent evidence on the safety of IBD drugs, cancer and recurrent cancer risk in patients with IBD and provides a framework for shared decision making involving patient, gastroenterologist and oncologist.

Differences in Willingness to Undergo BRCA1/2 Testing and Risk Reducing Surgery among the General Public, Cancer Patients, and Healthcare Professionals: A Large Population-Based Survey

We aimed to understand the decision-making process related to the willingness to undergo BRCA1/2 genetic testing, risk-reducing salpingo-oophorectomy (RRSO), or risk-reducing mastectomy (RRM) among the general public, cancer patients, and healthcare professionals in South Korea. In total, 3444 individuals (1496 from the general public, 1500 cancer patients, 108 clinicians, and 340 researchers) completed a survey addressing genetic testing and related risk management options in a hypothetical scenario. Differences in intent and associated factors for undergoing the above procedures or sharing test results were analyzed. Overall, 67% of participants were willing to undergo BRCA1/2 testing, with proportions of the general public (58%), cancer patients (70%), clinicians (88%), and researchers (90%). The willingness to undergo RRSO was highest among clinicians (58%), followed by among patients (38%), the general public (33%), and researchers (32%) (p < 0.001). Gender, age, education level, and household income were associated with willingness to undergo genetic testing, RRM, and RRSO (p < 0.05). The intent for undergo genetic testing, RRM, and RRSO were affected by many factors. Finally, 69% of the general public intended to share information with family, while this percentage was 92%, 91%, and 94% for patients, clinicians, and researchers, respectively (p < 0.05). These results highlight the requirement for developing targeted educational materials and counseling strategies for facilitating informed decision making.

What Does Your PROMIS Score Mean? Improving the Utility of Patient-Reported Outcomes at the Point of Care

STUDY DESIGN

Prospective cohort.

OBJECTIVES

Patient-Reported Outcome Measurement Information System (PROMIS) has been validated for lumbar spine. Use of patient-reported outcome (PRO) measures can improve clinical decision making and health literacy at the point of care. Use of PROMIS, however, has been limited in part because clinicians and patients lack plain language understanding of the meaning of scores and it remains unclear how best to use them at the point of care. The purpose was to develop plain language descriptions to apply to PROMIS Physical Function (PF) and Pain Interference (PI) scores and to assess patient understanding and preferences in presentation of their individualized PRO information.

METHODS

Retrospective analysis of prospectively collected PROMIS PF v1.2 and PI v1.1 for patients presenting to a tertiary spine center for back/lower extremity complaints was performed. Patients with missing scores, standard error >0.32, and assessments with <4 or >12 questions were excluded. Scores were categorized into score groups, specifically PROMIS PF groups were: <18, 20 ± 2, 25 ± 2, 30 ± 2, 35 ± 2, 40 ± 2, 45 ± 2, 50 ± 2, 55 ± 2, 60 ± 2, and >62; and PROMIS PI groups were: <48, 50 ± 2, 55 ± 2, 60 ± 2, 65 ± 2, 70 ± 2, 75 ± 2, 80 ± 2, and >82. Representative questions and answers from the PROMIS PI and PROMIS PF were selected for each score group, where questions with <25 assessments or representing <15% of assessments were excluded. Two fellowship-trained spine surgeons further trimmed the questions to create a streamlined clinical tool using a consensus process. Plain language descriptions for PROMIS PF were then used in a prospective assessment of 100 consecutive patients. Patient preference for consuming the score data was recorded and analyzed.

RESULTS

In total, 12 712 assessments/5524 unique patients were included for PF and 14 823 assessments/6582 unique patients for PI. More than 90% of assessments were completed in 4 questions. The number of assessments and patients per scoring group were normally distributed. The mean PF score was 37.2 ± 8.2 and the mean PI was 63.3 ± 7.4. Plain language descriptions and compact clinical tool was were generated. Prospectively 100 consecutive patients were surveyed for their preference in receiving their T-score versus plain language description versus graphical presentation. A total of 78% of patients found receiving personalized PRO data helpful, while only 1% found this specifically not helpful. Overall, 80% of patients found either graphical or plain language more helpful than T-score alone, and half of these preferred plain language and graphical descriptions together. In total, 89% of patients found the plain language descriptions to be accurate.

CONCLUSIONS

Patients at the point of care are interested in receiving the results of their PRO measures. Plain language descriptions of PROMIS scores enhance patient understanding of PROMIS numerical scores. Patients preferred plain language and/or graphical representation rather than a numerical score alone. While PROs are commonly used for assessing outcomes in research, use at point of care is a growing interest and this study clarifies how they might be utilized in physician-patient communication.

Making a Shared Decision on Meningococcal B Vaccine: Provider Feedback on an Educational Tool Developed for Use With Patients

BACKGROUND

In 2015, the Advisory Committee on Immunization Practices recommended Meningococcal B vaccination for healthy 16- to 23-year-olds based on shared clinical decision-making between the patient and the provider. There has been some confusion regarding how to implement this recommendation.

METHODS

Through discussions among the authors, a review of relevant literature, and consultation with vaccine experts, we developed educational materials for providers that included a patient handout to help initiate and guide conversations leading to shared clinical decision-making for the Meningococcal B vaccine. Materials were distributed to 88 health care providers who subsequently completed surveys to evaluate their impressions of the materials and the utility of the materials for clinical practice.

RESULTS

The survey results from the 88 providers revealed that they valued the materials; 93% percent indicated they would share these materials with colleagues, and 95% agreed or strongly agreed that they would share these materials with patient families. Responses to an open-ended question indicate that some providers initiated discussions regarding the Meningococcal B vaccine in ways that truncated conversation rather than encouraging a shared decision-making process.

CONCLUSIONS

Overall, the materials developed and implemented for this project support the initiation of, and help standardize provider conversations regarding, Meningococcal B vaccination for healthy adolescents.

Diagnostic accuracy of non-invasive tests for advanced fibrosis in patients with NAFLD: an individual patient data meta-analysis

OBJECTIVE

Liver biopsy is still needed for fibrosis staging in many patients with non-alcoholic fatty liver disease. The aims of this study were to evaluate the individual diagnostic performance of liver stiffness measurement by vibration controlled transient elastography (LSM-VCTE), Fibrosis-4 Index (FIB-4) and NAFLD (non-alcoholic fatty liver disease) Fibrosis Score (NFS) and to derive diagnostic strategies that could reduce the need for liver biopsies.

DESIGN

Individual patient data meta-analysis of studies evaluating LSM-VCTE against liver histology was conducted. FIB-4 and NFS were computed where possible. Sensitivity, specificity and area under the receiver operating curve (AUROC) were calculated. Biomarkers were assessed individually and in sequential combinations.

RESULTS

Data were included from 37 primary studies (n=5735; 45% women; median age: 54 years; median body mass index: 30 kg/m²; 33% had type 2 diabetes; 30% had advanced fibrosis). AUROCs of individual LSM-VCTE, FIB-4 and NFS for advanced fibrosis were 0.85, 0.76 and 0.73. Sequential combination of FIB-4 cut-offs (<1.3; ≥2.67) followed by LSM-VCTE cut-offs (<8.0; ≥10.0 kPa) to rule-in or rule-out advanced fibrosis had sensitivity and specificity (95% CI) of 66% (63-68) and 86% (84-87) with 33% needing a biopsy to establish a final diagnosis. FIB-4 cut-offs (<1.3; ≥3.48) followed by LSM cut-offs (<8.0; ≥20.0 kPa) to rule out advanced fibrosis or rule in cirrhosis had a sensitivity of 38% (37-39) and specificity of 90% (89-91) with 19% needing biopsy.

CONCLUSION

Sequential combinations of markers with a lower cut-off to rule-out advanced fibrosis and a higher cut-off to rule-in cirrhosis can reduce the need for liver biopsies.

Faecal immunochemical tests safely enhance rational use of resources during the assessment of suspected symptomatic colorectal cancer in primary care: systematic review and meta-analysis

OBJECTIVE

Implementation of faecal immunochemical tests (FIT) as a triage test in primary healthcare may improve the efficiency of referrals without missing cases of colorectal cancer (CRC). We aim to summarise the performance characteristics of FITs for CRC in symptomatic patients presenting to primary healthcare.

DESIGN

We performed a systematic literature review of Medline and EMBASE databases from May 2018 to November 2020. Previous related systematic searches were also adapted to this aim and completed with reference screening. We identified studies performed on adult patients consulting for abdominal symptoms in primary care which reported data such that the FIT diagnostic performance parameters for CRC could be obtained. Bivariate models were used to synthesise available evidence. Meta-regression analysis was performed to evaluate the causes of heterogeneity.

RESULTS

Twenty-three studies (69 536 participants) were included (CRC prevalence 0.3%-6.2%). Six studies (n=34 691) assessed FIT as rule in test (threshold of ≥150 µg Hb/g faeces) showing a sensitivity of 64.1% (95% CI 57.8% to 69.9%) and a specificity of 95.0% (95% CI 91.2% to 97.2%). A threshold of 10 µg/g (15 studies; n=48 872) resulted in a sensitivity of 87.2% (95% CI 81.0% to 91.6%) and a specificity of 84.4% (95% CI 79.4% to 88.3%) for CRC. At a 20 µg Hb/g faeces threshold (five studies; n=24 187) less than one additional CRC would be missed per 1000 patients investigated compared with 10 µg Hb/g faeces threshold (CRC prevalence 2%).

CONCLUSION

FIT is the test of choice to evaluate patients with new-onset lower gastrointestinal symptoms in primary healthcare.

Treat-to-target in axial spondyloarthritis: an observational study in daily practice

OBJECTIVES

To evaluate the extent to which internationally agreed treat-to-target recommendations were applied in clinical practice in patients with axial spondyloarthritis.

METHODS

Data were used from a web-based patient registry for monitoring SpA in daily practice in the Netherlands. The extent to which treat-to-target was applied was evaluated through four indicators: the proportion of patients (i) with ≥1 Ankylosing Spondylitis Disease Activity Score (ASDAS) assessed during a 1-year period, (ii) having inactive disease/low disease activity (i.e. ASDAS < 2.1), (iii) in whom re-evaluation of ASDAS within recommended intervals occurred, and (iv) with high disease activity (HDA, i.e. ASDAS ≥ 2.1) in whom treatment was adapted ≤6 weeks after obtaining ASDAS ≥ 2.1. Patients with HDA with treatment adaptations were compared with patients with HDA without treatment adaptations.

RESULTS

In 185 out of 219 patients (84%), disease activity was monitored with ≥1 ASDAS during a 1-year period, of whom 71 (38%) patients had a score below the target (ASDAS < 2.1) at first measurement. Re-evaluation of ASDAS ≤3 months occurred in 11% and 23% of the patients with inactive disease/low disease activity and HDA, respectively. Treatment adaptation occurred in 19 out of 114 patients (17%) with HDA. Patients in whom treatment was adapted had significantly higher ASDAS (P < 0.01), CRP levels (P < 0.05) and physician global assessment (P < 0.05) compared with patients without treatment adaptations.

CONCLUSIONS

Treat-to-target was applied to a limited extent in clinical practice in patients with axial spondyloarthritis. Available disease activity scores seemed not to be used for determining the frequency of re-evaluation nor treatment adaptation.

Optimal dynamic treatment regime estimation using information extraction from unstructured clinical text

The wide-scale adoption of electronic health records (EHRs) provides extensive information to support precision medicine and personalized health care. In addition to structured EHRs, we leverage free-text clinical information extraction (IE) techniques to estimate optimal dynamic treatment regimes (DTRs), a sequence of decision rules that dictate how to individualize treatments to patients based on treatment and covariate history. The proposed IE of patient characteristics closely resembles "The clinical Text Analysis and Knowledge Extraction System" and employs named entity recognition, boundary detection, and negation annotation. It also utilizes regular expressions to extract numerical information. Combining the proposed IE with optimal DTR estimation, we extract derived patient characteristics and use tree-based reinforcement learning (T-RL) to estimate multistage optimal DTRs. IE significantly improved the estimation in counterfactual outcome models compared to using structured EHR data alone, which often include incomplete data, data entry errors, and other potentially unobserved risk factors. Moreover, including IE in optimal DTR estimation provides larger study cohorts and a broader pool of candidate tailoring variables. We demonstrate the performance of our proposed method via simulations and an application using clinical records to guide blood pressure control treatments among critically ill patients with severe acute hypertension. This joint estimation approach improves the accuracy of identifying the optimal treatment sequence by 14-24% compared to traditional inference without using IE, based on our simulations over various scenarios. In the blood pressure control application, we successfully extracted significant blood pressure predictors that are unobserved or partially missing from structured EHR.

Multicentre derivation and validation of a colitis-associated colorectal cancer risk prediction web tool

OBJECTIVE

Patients with ulcerative colitis (UC) diagnosed with low-grade dysplasia (LGD) have increased risk of developing advanced neoplasia (AN: high-grade dysplasia or colorectal cancer). We aimed to develop and validate a predictor of AN risk in patients with UC with LGD and create a visual web tool to effectively communicate the risk.

DESIGN

In our retrospective multicentre validated cohort study, adult patients with UC with an index diagnosis of LGD, identified from four UK centres between 2001 and 2019, were followed until progression to AN. In the discovery cohort (n=246), a multivariate risk prediction model was derived from clinicopathological features using Cox regression. Validation used data from three external centres (n=198). The validated model was embedded in a web tool to calculate patient-specific risk.

RESULTS

Four clinicopathological variables were significantly associated with AN progression in the discovery cohort: endoscopically visible LGD >1 cm (HR 2.7; 95% CI 1.2 to 5.9), unresectable or incomplete endoscopic resection (HR 3.4; 95% CI 1.6 to 7.4), moderate/severe histological inflammation within 5 years of LGD diagnosis (HR 3.1; 95% CI 1.5 to 6.7) and multifocality (HR 2.9; 95% CI 1.3 to 6.2). In the validation cohort, this four-variable model accurately predicted future AN cases with overall calibration Observed/Expected=1.01 (95% CI 0.64 to 1.52), and achieved 100% specificity for the lowest risk group over 13 years of available follow-up.

CONCLUSION

Multicohort validation confirms that patients with large, unresected, multifocal LGD and recent moderate/severe inflammation are at highest risk of developing AN. Personalised risk prediction provided via the Ulcerative Colitis-Cancer Risk Estimator ( www.UC-CaRE.uk ) can support treatment decision-making.

Leveraging Large-Scale Electronic Health Records and Interpretable Machine Learning for Clinical Decision Making at the Emergency Department: Protocol for System Development and Validation

Background

There is a growing demand globally for emergency department (ED) services. An increase in ED visits has resulted in overcrowding and longer waiting times. The triage process plays a crucial role in assessing and stratifying patients’ risks and ensuring that the critically ill promptly receive appropriate priority and emergency treatment. A substantial amount of research has been conducted on the use of machine learning tools to construct triage and risk prediction models; however, the black box nature of these models has limited their clinical application and interpretation.

Objective

In this study, we plan to develop an innovative, dynamic, and interpretable System for Emergency Risk Triage (SERT) for risk stratification in the ED by leveraging large-scale electronic health records (EHRs) and machine learning.

Methods

To achieve this objective, we will conduct a retrospective, single-center study based on a large, longitudinal data set obtained from the EHRs of the largest tertiary hospital in Singapore. Study outcomes include adverse events experienced by patients, such as the need for an intensive care unit and inpatient death. With preidentified candidate variables drawn from expert opinions and relevant literature, we will apply an interpretable machine learning–based AutoScore to develop 3 SERT scores. These 3 scores can be used at different times in the ED, that is, on arrival, during ED stay, and at admission. Furthermore, we will compare our novel SERT scores with established clinical scores and previously described black box machine learning models as baselines. Receiver operating characteristic analysis will be conducted on the testing cohorts for performance evaluation.

Results

The study is currently being conducted. The extracted data indicate approximately 1.8 million ED visits by over 810,000 unique patients. Modelling results are expected to be published in 2022.

Conclusions

The SERT scoring system proposed in this study will be unique and innovative because of its dynamic nature and modelling transparency. If successfully validated, our proposed solution will establish a standard for data processing and modelling by taking advantage of large-scale EHRs and interpretable machine learning tools.

International Registered Report Identifier (IRRID)

DERR1-10.2196/34201

A Concept Analysis of Nurses' Clinical Decision Making: Implications for Korea

The study's purpose was to identify the meaning and the attributes of Korean nurses' clinical decision making. A sequential and systematic literature review with reflection according to the conceptual analysis method of Walker and Avant was used in this study. Data sources included the National Assembly Library, the National Digital Science Library, ProQuest, PubMed, MEDLINE, and CINAHL. Finally, twenty-six articles were included in this concept analysis. The concept of Korean nurses' clinical decision making consisted of the following attributes: clinical reasoning, choosing and applying challenging alternatives, and professional assessment and resetting. Antecedents consisted of: recognizing complex and diverse patient situations with high uncertainty, the need to solve problems according to priority, prior experience in clinical decision making, and interrelationships with fellow medical staff. Consequences consisted of: providing high-quality nursing services, improving the patient's safety, and increased satisfaction with clinical decision making. Based on these results, the conceptual attributes of Korean nurses' clinical decision making had slightly different characteristics but were organically interrelated. The results of analyzing the concept of Korean nurses' clinical decision making provide a better understanding of it and contribute to expanding nursing knowledge and developing a valid and reliable measurement.

Thrombotic Microangiopathy Due to Malignant Hypertension Treated Exclusively With Antihypertensive Therapy

Plasma exchange is the commonly considered therapy for the treatment of thrombotic microangiopathy (TMA); however, it is not always mandatory. We treated a patient who presented with malignant hypertension (MH) complicated by TMA using antihypertensive therapy that was not accompanied by plasma exchange. A 38-year-old woman with photophobia, diarrhea, fever, and severely elevated blood pressure was referred to our hospital. Blood test results revealed thrombocytopenia and hemolytic anemia, and ascites were observed on the computed tomography images. Although TMA was suspected, plasma exchange was not performed because the platelet count was not markedly low. Her blood cell counts improved after antihypertensive treatment, and she was discharged. The patient is currently under therapy and remains stable. Thus, TMA secondary to MH may improve using antihypertensive therapy, without the need for invasive plasma exchange. Considering the platelet count may be helpful in deciding whether plasma exchange is required.