Diseases Codes Research Articles

ACUTE PANCREATITIS IN AKMOLA REGION ACCORDING TO EMERGENCY SURGERY

Background: Acute pancreatitis is a major issue in surgical practice, with high complications and mortality rates. It severely affects patient quality of life and places a significant economic strain on healthcare systems, making its study crucial for public health. This study aims to analyze trends in acute pancreatitis incidence in the Akmola region from 2009 to 2022. Materials and methods: The study is based on a retrospective analysis of data from the Ministry of Health of the Republic of Kazakhstan according to the International Classification of Diseases-10 code: K85. Methods of health statistics and time series analysis were used to assess the dynamics of morbidity. Results: There was an increase in the total number of hospitalizations from 865 cases in 2009 to 1.061 in 2022. The percentage of surgical interventions decreased from 5.7% to 4.2%, while the share of conservative treatment increased to 95.8%. The mortality rate among operated patients was 20.4% in 2019 and increased to 33.3% in 2022. The incidence of hospitalization increased from 117.1 per 100000 population in 2009 to 144.6 per 100000 population in 2022, with a decrease in the rate of late visits from 42.6 to 36.1 per 100000 population, respectively. Conclusion: It has been established that for patients in the second group according to the Enhanced Recovery After Surgery strategy, hemodynamic support and effective pain management can contribute to early patient mobilization after surgery. Early mobilization, in turn, can expedite recovery and reduce the length of hospital stay, ultimately leading to potential cost savings. Conclusion. The study highlights the need to improve medical approaches to acute pancreatitis in the Akmola region. The results indicate the critical importance of early diagnosis and timely hospitalization to reduce complications and mortality. Additional research is needed to develop more effective clinical protocols.

Effects of Social Vulnerability and Environmental Burden on Care Fragmentation and Social Needs Among Individuals With Rheumatic Conditions.

Environmental hazards and heightened neighborhood social vulnerability coexist and disproportionately affect minoritized populations. We investigated associations between exposure to adverse environmental burden concentrated in areas with high social vulnerability and care fragmentation (missed appointments, emergency department visits, and hospitalizations) and social needs (eg, food and housing insecurity) among individuals with rheumatic conditions. We identified adults receiving care in a Massachusetts multihospital system with at least two rheumatic disease codes and complete street addresses. Geocoded addresses were linked to the Centers for Disease Control and Prevention/Agency for Toxic Substances and Disease Registry Social-Environmental Ranking (SER), which combines census-tract social vulnerability variables (eg, socioeconomic status) with environmental hazards (eg, air and water pollution). Social needs were obtained from self-reported surveys. Multilevel, multinomial regression models estimated associations between SER quartiles and care fragmentation and social need burden, accounting for demographics and comorbidities. Among 16,856 individuals with rheumatic conditions, 70% were female, 6% were Black, 82% were White, and 7% resided in the highest combined social vulnerability and environmental burden (SER quartile 4) areas. Among 7,083 with social needs data, 19% experienced more than one challenge. Individuals in SER quartile 4 areas (vs quartile 1) had 2.02 (95% confidence interval [CI] 1.67-2.46) times greater odds of at least four care fragmentation occurrences (vs 0) and 2.37 (95% CI 1.73-3.25) times greater odds of at least two social needs (vs 0). Residence in areas of high combined adverse environmental burden and social vulnerability was associated with significantly greater odds of care fragmentation and social needs. Addressing structural factors and emerging environmental threats contributing to these adverse exposures is essential to reduce rheumatic disease care inequities.

Atrial flutter and sick sinus syndrome

BackgroundSick sinus syndrome (SSS) is a common condition resulting in reduced quality of life, syncope, and permanent pacemaker (PPM) implantation, but predictors have not been elucidated. Whereas atrial arrhythmias are frequently associated with SSS, we hypothesized that atrial flutter (AFL) would strongly predict SSS, given shared relationships with right atrial and particularly crista terminalis fibrosis. ObjectiveThe study aimed to assess the impact of AFL on the occurrence of SSS and associated syncope and PPM implantation. MethodsHealth care databases were used to identify adults aged ≥18 years receiving hospital-based care in California in 2005–2019. International Classification of Diseases codes were used to identify diagnoses and procedures. Patients were classified on the basis of the presence of AFL and atrial fibrillation (AF). Cox proportional hazards models adjusting for demographics and comorbidities were employed. ResultsWe included 29,357,609 individuals (54% female; mean age, 46 years), 101,243 with AFL alone, 1,674,680 with AF alone, and 284,547 with AF and AFL. After adjustment for age, sex, race and ethnicity, and comorbidities, AF, AFL, and both arrhythmias were each associated with increased risk of SSS and associated syncope and PPM implantation (all P < .001). In the population with AF, an additional AFL diagnosis conferred a higher risk for development of SSS (hazard ratio [HR],1.62; 95% confidence interval, 1.59–1.64), syncope (HR, 1.63; 1.54–1.72), and PPM implantation (HR, 1.74; 1.70–1.79). ConclusionAFL is associated with an increased risk of incident SSS and its adverse consequences, especially in patients with coexisting AF. AFL may be useful for risk stratification strategies to predict, to prevent, and to treat SSS.

PheW2P2V: a phenome-wide prediction framework with weighted patient representations using electronic health records.

Electronic health records (EHRs) provide opportunities for the development of computable predictive tools. Conventional machine learning methods and deep learning methods have been widely used for this task, with the approach of usually designing one tool for one clinical outcome. Here we developed PheW2P2V, a Phenome-Wide prediction framework using Weighted Patient Vectors. PheW2P2V conducts tailored predictions for phenome-wide phenotypes using numeric representations of patients' past medical records weighted based on their similarities with individual phenotypes. PheW2P2V defines clinical disease phenotypes using Phecode mapping based on International Classification of Disease codes, which reduces redundancy and case-control misclassification in real-life EHR datasets. Through upweighting medical records of patients that are more relevant to a phenotype of interest in calculating patient vectors, PheW2P2V achieves tailored incidence risk prediction of a phenotype. The calculation of weighted patient vectors is computationally efficient, and the weighting mechanism ensures tailored predictions across the phenome. We evaluated prediction performance of PheW2P2V and baseline methods with simulation studies and clinical applications using the MIMIC-III database. Across 942 phenome-wide predictions using the MIMIC-III database, PheW2P2V has median area under the receiver operating characteristic curve (AUC-ROC) 0.74 (baseline methods have values ≤0.72), median max F1-score 0.20 (baseline methods have values ≤0.19), and median area under the precision-recall curve (AUC-PR) 0.10 (baseline methods have values ≤0.10). PheW2P2V can predict phenotypes efficiently by using medical concept embeddings and upweighting relevant past medical histories. By leveraging both labeled and unlabeled data, PheW2P2V reduces overfitting and improves predictions for rare phenotypes, making it a useful screening tool for early diagnosis of high-risk conditions, though further research is needed to assess the transferability of embeddings across different databases. PheW2P2V is fast, flexible, and has superior prediction performance for many clinical disease phenotypes across the phenome of the MIMIC-III database compared to that of several popular baseline methods.

Causal relationship between circulating immune cells and gastric cancer: a bidirectional Mendelian randomization analysis using UK Biobank and FinnGen datasets.

The role of immune cells in cancer pathogenesis remains controversial due to conflicting reports, potentially arising from various confounding factors. Emerging evidence suggests that cancer can also influence immune cell populations and functions, making it challenging to investigate their causal relationship. Traditional observational studies often fail to eliminate all confounding factors and are prone to reverse causality. Therefore, we employ Mendelian randomization (MR) to determine the causal relationship between immune cells and cancer, as this method can identify causal relationships independent of confounding factors and avoid reverse causality. Genome-wide association study (GWAS) summary statistics on immune traits, encompassing 310 immune cell phenotypes, were obtained from 3,757 European individuals, with peripheral blood immune cells tested using flow cytometry. GWAS summary statistics for gastric cancer were derived from 476,116 European individuals across two large-scale biobanks: the UK Biobank and FinnGen. Gastric cancer was identified by the International Classification of Diseases, 9th Revision (ICD-9), and 10th Revision (ICD-10) codes. Significant single nucleotide polymorphisms (SNPs) for immune traits were extracted at a threshold of P<1×10-5, while a threshold of P<5×10-8 was used for gastric cancer GWAS data. Linkage imbalance-based clumping was performed to obtain independent SNPs, and those with F<10 were excluded to mitigate weak instrument bias. Phenoscanner V2 was used to exclude SNPs directly associated with potential confounders or outcomes. Two-sample MR was conducted using five MR methods, with inverse-variance-weighted (IVW) as the primary analysis method. A false discovery rate (FDR) correction was used to reduce the likelihood of type 1 errors. In addition, we conducted MR-Egger intercept tests and Cochran's Q tests. The numbers of CD4-CD8- T cells and IgD-CD27- B cells were positively correlated with the development of gastric cancer, with odds ratios (ORs) of 1.15 [95% confidence interval (CI), 1.07-1.24; P<0.001; PFDR=0.041; IVW method] and 1.07 (95% CI, 1.03-1.11; P=0.001; PFDR=0.187; IVW method), respectively. However, the percentage of IgD+CD24- B cells in lymphocytes were negatively associated with the development of gastric cancer (OR =0.90; 95% CI, 0.84-0.96; P=0.002; PFDR=0.187; IVW method). MR analysis of the above three immune cell phenotypes showed no significant heterogeneity or horizontal pleiotropy. In the reverse MR analysis, gastric cancer was not causally associated with any of the immune cell phenotypes. Circulating CD4-CD8- T cells and IgD-CD27- B cells are positively correlated with the development of gastric cancer, while the percentage of IgD+CD24- B cells in lymphocytes are negatively correlated. These findings provide insight into the relationship between immune cells and gastric cancer pathogenesis and may serve as a basis for the development of immunotherapies for gastric cancer.

Association Between Admission Systolic Blood Pressure and Outcomes in Patients with Isolated Traumatic Brain Injury: A Cross-National Multicenter Cohort Study.

The optimal prehospital blood pressure in patients following traumatic brain injury (TBI) remains controversial. We aimed to assess the association between the systolic blood pressure (SBP) at emergency department triage and patient outcomes following isolated moderate-to-severe TBI. We conducted a cross-national multicenter retrospective cohort study using the Pan-Asia Trauma Outcomes Study database from January 1, 2016, to November 30, 2018. The enrollees were adult patients with isolated moderate-to-severe TBI defined by the International Classification of Diseases code, a Glasgow Coma Scale (GCS) <13 at triage, and a nonhead Abbreviated Injury Scale ≤3. The studied variables were SBPs at triage categorized into different ranges. The primary outcome was 30-day mortality, and the secondary outcome was poor functional status at hospital discharge defined by the modified Rankin Scale ≥4. Multivariable logistic regression was applied to adjust for confounders including country, sex, age, mechanism of injury, prehospital vascular access, respiratory rate, GCS, oxygen saturation, intubation, Injury Severity Score, head surgery, intensive care unit admission, and length of hospital stay. Subgroup analyses were performed on different severity of TBI. A total of 785 patients (median age, 42 years; male patients 77.5%; mean SBP at triage, 136.3 ± 33.1 mmHg) were included in the primary analysis. The lowest 30-day mortality rate existed in patients with SBP of 100-119 mmHg. Taking it as baseline, the adjusted odds ratios (aORs) and 95% confidence intervals (CIs) of SBP <100 mmHg, 120-139 mmHg, 140-159 mmHg, and ≥160 mmHg were 7.05 (2.51-19.78), 3.14 (1.14-8.65), 2.91 (1.04-8.17), and 3.28 (1.14-9.42). As for the secondary outcome, the aORs and 95% CIs were 1.36 (0.68-2.68) of <100 mmHg, 0.99 (0.57-1.70) of 120-139 mmHg, 1.23 (0.67-2.25) of 140-159 mmHg, and 1.52 (0.78-2.95) of ≥160 mmHg. Subgroup analyses revealed trends of the best outcomes in both moderate and severe TBI patients with SBP 100-119 mmHg, whereas statistical significance appeared only in patients with severe TBI. SBP of 110-119 mmHg at triage is associated with the lowest 30-day mortality in patients following isolated moderate-to-severe TBI and possibly related to a better functional outcome.

Machine learning prediction of pancreatitis risk with antithyroid drugs: a nationwide retrospective observational study.

In recent years, there has been increasing data showing that the risk of acute pancreatitis (AP) is increased in patients using methimazole (MMI). The aim of this population-based study was to investigate the association between drugs used in the treatment of hyperthyroidism (MMI, PTU, propranolol) and the diagnosis of AP. The database consisted of more than 85 million citizen records between January 1, 2017 and December 31, 2022. Within the scope of the study, the data were examined of patients diagnosed with hyperthyroidism and the individuals who developed acute pancreatitis after being diagnosed with hyperthyroidism, which was accomplished using the International Classification of Diseases-10 (ICD-10) codes. Analyses were carried out using R Studio 4.2.0 software. The test-train technique as a machine learning method was used in cox regression models. A total of 1,329,934 individuals diagnosed with hyperthyroidism were examined. Of these, 10,815 (0.81%) patients developed acute pancreatitis after being diagnosed with hyperthyroidism. The use of propylthiouracil was determined to increase the probability of experiencing AP diagnosis 0.10-fold (10%) (p:0.0009), and the use of propranolol decrease the probability of experiencing AP 0.5-fold (5%) (p:0.032), with a 52.2% predictive success rate. There is no any statistically significant association between MMI use and the probability of experiencing AP according to the model have been build. Although the risk of experiencing AP was not increased by MMI use, PTU use increased and propronolol use decrease the risk of experiencing AP in patients with hyperthyroidism. Although drug-induced acute pancreatitis is relatively rare, a firm understanding of the drugs associated with the condition should alert the clinician to appropriately diagnose and treat patients.

Contemporary nationwide trends in major adverse cardiovascular events in young cannabis users without concomitant tobacco, alcohol, cocaine use.

Cannabis use has increased among young individuals in recent years. Although dependent cannabis use disorder (CUD) has been associated with various cardiac events, its effects on young adults without concurrent substance use remain understudied. To examine trends in hospitalizations for major adverse cardiac and cerebrovascular events (MACCE) in this cohort. We used the National Inpatient Sample (2016-2019) to identify hospitalized young individuals (18-44 years), excluding those with concurrent substance usage (tobacco, alcohol, and cocaine). They were divided into CUD+ and CUD-. Using International Classification of Diseases-10 codes, we examined the trends in MACCE hospitalizations, including all-cause mortality (ACM), acute myocardial infarction (AMI), cardiac arrest (CA), and acute ischemic stroke (AIS). Of 27.4 million hospitalizations among young adults without concurrent substance abuse, 4.2% (1.1 million) had co-existent CUD. In CUD+ group, hospitalization rates for MACCE (1.71% vs 1.35%), AMI (0.86% vs 0.54%), CA (0.27% vs 0.24%), and AIS (0.49% vs 0.35%) were higher than in CUD- group (P < 0.001). However, rate of ACM hospitalizations was lower in CUD+ group (0.30% vs 0.44%). From 2016 to 2019, CUD+ group experienced a relative rise of 5% in MACCE and 20% in AMI hospitalizations, compared to 22% and 36% increases in CUD- group (P < 0.05). The CUD+ group had a 13% relative decrease in ACM hospitalizations, compared to a 10% relative rise in CUD- group (P < 0.05). However, when adjusted for confounders, MACCE odds among CUD+ cohort remain comparable between 2016 and 2019. The CUD+ group had higher rates of MACCE, but the rising trends were more apparent in the CUD- group over time. Interestingly, the CUD+ group had lower ACM rates than the CUD- group.

Building a National Policy for Rare Disease in Brazil.

Rare diseases (RD) are individually rare, although encompass a significant proportion of the population, affecting not only the individuals but also their families. In Brazil RD is defined by the Ministry of Health as a disorder that affects up to 65 individuals in 100,000, or 1.3 individuals in every 2,000. In this review the environment that led to the publication of a National Policy for Comprehensive Care for People of Rare Disease in 2014, a national plan with the aim to decrease morbidity and mortality of RD, improving the care of people with RD in the public health system are described. The process that finally led to such policy took over a decade, moving forward not only due to technical needs, but having patient organizations as essential actors and advocates. Specialized centers in RD were licensed and, since its publication, 33 centers have been accredited; such process, however, has been slow and concentrated in specific regions and larger cities of the country. Despite the incorporation of genetic tests in 2014 and exome sequencing later in 2020, many genetic tests are not offered by specialized centers, with unequal availability across the country. Public health system in Brazil uses ICD-10 for disease coding, preventing appropriate epidemiologic knowledge of RD in Brazil. Incorporation of new technologies as orphan drugs has been in place and regulation for expedite licensing for new RD drugs were issued, although high cost and availability to RD population has been a challenge.

Dermatologic Conditions in Down Syndrome: A Multi-Site Retrospective Review of International Classification of Diseases Codes.

As the population and life expectancy of people with Down syndrome increases, identifying common skin conditions throughout the lifespan will help inform clinical care and research. We sought to evaluate dermatologic conditions diagnosed in pediatric and adult patients with Down syndrome. This multicenter retrospective study analyzed demographic and outpatient visit International Classification of Diseases codes of patients with Down syndrome evaluated at outpatient dermatology clinics in the United States or Canada between 2011 and 2021. A total of 1529 patients with Down syndrome were identified from eight academic medical centers: 50.8% were children (0-12 years), 25.2% were adolescents (13-17 years), and 24% were adults (18 years and older). Eczematous dermatitis was the most common diagnosis overall (26%), followed by folliculitis (19.3%) and seborrheic dermatitis (15.6%). Other notable diagnoses included dermatophyte infections (13%), alopecia areata (11.6%), and psoriasis (6.7%). About 4.3% of visits included a code for high-risk medication use. Eczematous dermatitis, alopecia areata, and folliculitis were the most common diagnoses observed in children; folliculitis, hidradenitis suppurativa, and eczematous dermatitis in adolescents; and seborrheic dermatitis, eczematous dermatitis, and folliculitis in adults. Dermatologic conditions in patients with Down syndrome vary by age, but are most often eczematous, adnexal, and cutaneous autoimmune disorders. This multicenter retrospective review identifies skin diseases that should be prioritized for clinical care guideline development and research in the Down syndrome community.

Nontobacco Nicotine Dependence and Rates of Periprosthetic Joint Infection and Other Postoperative Complications in Shoulder Arthroplasty: A Retrospective Analysis.

Nontobacco nicotine products, including e-cigarettes and vaping, are marketed as healthier alternatives to tobacco. However, the literature on nontobacco nicotine dependence (NTND) is scarce. It is important to analyze the influence of these products as they pertain to medical and surgical postoperative complications. This study hypothesizes that patients with NTND will experience more postoperative complications. Using the TriNetX database, which aggregates deidentified medical records from 89 healthcare organizations in the Research Network, Current Procedural Terminology and 10th revision of the International Classification of Diseases codes were used to identify patients undergoing primary shoulder arthroplasty (SA) from January 2012 to February 2024. Patients were divided into cohorts based on their NTND status before surgery. 90-day major medical complications and 2-year implant-related complications were assessed. Statistical analyses involved calculating risk ratios for postoperative complications. This study analyzed a total of 89,910 SA patients, of which 6,756 were 1:1 propensity matched into NTND or control cohorts. Within the 90-day postoperative period, the NTND cohort exhibited significantly higher rates of sepsis (1.80 vs. 1.20, P = 0.012), surgical site infection (1.20 vs. 0.70%, P = 0.007), and wound disruptions (0.70 vs. 0.40%, P = 0.048), average opioids prescribed (4.46 vs. 3.338, P < 0.001), readmission (10.20% vs. 6.20%, P 0.001) compared with the non-NTND cohort. At the 2-year follow-up, mechanical loosening was notably higher in the NTND group (1.10 vs. 0.30%, P 0.001), as were rates of prosthetic joint infections (2.20 vs. 1.20%, P 0.001). No significant difference was observed for revision rates (3.20% vs. 2.90%, P = 0.269). NTND is associated with higher 90-day rates of wound distruptions, infections, sepsis, as well as increased rates of mechanical loosening and prosthetic joint infection at 2 years postoperatively after SA. These results highlight the need for comprehensive NTND preoperative screening and tailored patient counseling in this patient population.

Comprehensive Codified Algorithms to Identify the Underestimated Burden of Hidradenitis Suppurativa in the United States.

Hidradenitis suppurativa (HS) is a painful, inflammatory skin disease associated with a high disease burden and long diagnostic delay. Prevalence estimates of HS vary widely in the literature owing to differing estimation methodologies. This study aimed to apply stepwise algorithms to estimate the prevalence of possible/diagnosed cases of HS in the US. This was a retrospective cohort study in adult and pediatric patients with HS which utilized data from four US databases (MarketScan [Medicare and Medicaid] andOptum [electronichealth record (EHR) andClinformatics Data Mart (CDM)]). Patients withpossible/diagnosed HS were identified using two algorithms (termed Algorithm 1 and Algorithm 2), which assessed symptoms such as multiple skin boils in site-specific areas based on international classification of disease (ICD) codes. Patients with diagnosed HS were defined as having ≥ 2 outpatient or ≥ 1 inpatient diagnosis codes of HS. In each database, patients with continuous medical and pharmacy benefits in the 365days pre-index and 0-365days post-index periods were eligible for inclusion. Across all databases, Algorithm 2 (MarketScan Medicare [N = 309,916]; MarketScan Medicaid [N = 188,783]; Optum EHR [N = 366,158]; Optum CDM [N = 173,812]) identified more patients with possible/diagnosed HS than Algorithm 1 (MarketScan Medicare [N = 194,353]; MarketScan Medicaid [N = 99,276]; Optum EHR [N = 177,957]; Optum CDM [N = 112,244]). Based on ICD-9/10 codes, the 5-year period prevalence of HS ranged from 0.06% to 0.12% across all databases, while for Algorithm 1 and Algorithm 2, this ranged from 0.27% to 0.41% and 0.49% to 0.78%, respectively. Adults and females generally had a higher 5-year period prevalence versus pediatric patients and males, respectively. This real-world study highlights that HS diagnosis codes alone may be insufficient to estimate the prevalence of HS, demonstrating the value of employing algorithms in practice which assess for parameters such as multiple skin boils in site-specific areas. Integrating robust methods to identify the prevalence of HS may improve the diagnostic delay observed in HS and improve treatment outcomes.

Machine learning approach to predict acute kidney injury among patients undergoing multi-level spinal posterior instrumented fusion.

Acute kidney injury (AKI) after spinal fusion is a significant morbidity that can lead to poor post-surgical outcomes. Identifying AKI risk factors and developing a risk model can raise surgeons' awareness and allow them to take actions to mitigate the risks. The objective of the current study is to develop machine learning (ML) models to assess patient risk factors predisposing to AKI after posterior spinal instrumented fusion. Data was collected from the IBM MarketScan Database (2009-2021) for patients >18 years old who underwent spinal fusion with posterior instrumentation (3-6 levels). AKI incidence (defined by the International Classification of Diseases codes) was recorded 90-day post-surgery. Risk factors for AKI were investigated and compared through several ML models including logistic regression, linear support vector machine (LSVM), random forest, extreme gradient boosting (XGBoost), and neural networks. Among the 141,697 patients who underwent fusion with posterior instrumentation (3-6 levels), the overall rate of 90-day AKI was 2.96%. We discovered that the logistic regression model and LSVM demonstrated the best predictions with area under the curve (AUC) values of 0.75. The most important AKI prediction features included chronic renal disease, hypertension, diabetes mellitus ± complications, older age (>50 years old), and congestive heart failure. Patients who did not have these five key risk factors had a 90-day AKI rate of 0.29%. Patients who had an increasing number of key risk factors subsequently had higher risks of postoperative AKI. The analysis of the data with different ML models identified 5 key variables that are most closely associated with AKI: chronic renal disease, hypertension, diabetes mellitus ± complications, older age (>50 years old), and congestive heart failure. These variables constitute a simple risk calculator with additive odds ratio ranging from 3.38 (1 risk factor) to 91.10 (5 risk factors) over 90 days after posterior spinal fusion surgery. These findings can help surgeons risk-stratify their patients for AKI risk, and potentially guide post-operative monitoring and medical management.

Evaluating a Natural Language Processing-Driven, AI-Assisted International Classification of Diseases, 10th Revision, Clinical Modification, Coding System for Diagnosis Related Groups in a Real Hospital Environment: Algorithm Development and Validation Study.

International Classification of Diseases codes are widely used to describe diagnosis information, but manual coding relies heavily on human interpretation, which can be expensive, time consuming, and prone to errors. With the transition from the International Classification of Diseases, Ninth Revision, to the International Classification of Diseases, Tenth Revision (ICD-10), the coding process has become more complex, highlighting the need for automated approaches to enhance coding efficiency and accuracy. Inaccurate coding can result in substantial financial losses for hospitals, and a precise assessment of outcomes generated by a natural language processing (NLP)-driven autocoding system thus assumes a critical role in safeguarding the accuracy of the Taiwan diagnosis related groups (Tw-DRGs). This study aims to evaluate the feasibility of applying an International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM), autocoding system that can automatically determine diagnoses and codes based on free-text discharge summaries to facilitate the assessment of Tw-DRGs, specifically principal diagnosis and major diagnostic categories (MDCs). By using the patient discharge summaries from Kaohsiung Medical University Chung-Ho Memorial Hospital (KMUCHH) from April 2019 to December 2020 as a reference data set we developed artificial intelligence (AI)-assisted ICD-10-CM coding systems based on deep learning models. We constructed a web-based user interface for the AI-assisted coding system and deployed the system to the workflow of the certified coding specialists (CCSs) of KMUCHH. The data used for the assessment of Tw-DRGs were manually curated by a CCS with the principal diagnosis and MDC was determined from discharge summaries collected at KMUCHH from February 2023 to April 2023. Both the reference data set and real hospital data were used to assess performance in determining ICD-10-CM coding, principal diagnosis, and MDC for Tw-DRGs. Among all methods, the GPT-2 (OpenAI)-based model achieved the highest F1-score, 0.667 (F1-score 0.851 for the top 50 codes), on the KMUCHH test set and a slightly lower F1-score, 0.621, in real hospital data. Cohen κ evaluation for the agreement of MDC between the models and the CCS revealed that the overall average κ value for GPT-2 (κ=0.714) was approximately 12.2 percentage points higher than that of the hierarchy attention network (κ=0.592). GPT-2 demonstrated superior agreement with the CCS across 6 categories of MDC, with an average κ value of approximately 0.869 (SD 0.033), underscoring the effectiveness of the developed AI-assisted coding system in supporting the work of CCSs. An NLP-driven AI-assisted coding system can assist CCSs in ICD-10-CM coding by offering coding references via a user interface, demonstrating the potential to reduce the manual workload and expedite Tw-DRG assessment. Consistency in performance affirmed the effectiveness of the system in supporting CCSs in ICD-10-CM coding and the judgment of Tw-DRGs.

Real-world data to support post-market safety and performance of embolization coils: evidence generation from a medical device manufacturer and data institute partnership

BackgroundRecognizing the limitations of pre-market clinical data, regulatory authorities have embraced total product lifecycle management with post-market surveillance (PMS) data to assess medical device safety and performance. One method of proactive PMS involves the analysis of real-world data (RWD) through retrospective review of electronic health records (EHR). Because EHRs are patient-centered and focused on providing tools that clinicians use to determine care rather than collecting information on individual medical products, the process of transforming RWD into real-world evidence (RWE) can be laborious, particularly for medical devices with broad clinical use and extended clinical follow-up. This study describes a method to extract RWD from EHR to generate RWE on the safety and performance of embolization coils.MethodsThrough a partnership between a non-profit data institute and a medical device manufacturer, information on implantable embolization coils’ use was extracted, linked, and analyzed from clinical data housed in an electronic data warehouse from the state of Indiana’s largest health system. To evaluate the performance and safety of the embolization coils, technical success and safety were defined as per the Society of Interventional Radiology guidelines. A multi-prong strategy including electronic and manual review of unstructured (clinical chart notes) and structured data (International Classification of Disease codes), was developed to identify patients with relevant devices and extract data related to the endpoints.ResultsA total of 323 patients were identified as treated using Cook Medical Tornado, Nester, or MReye embolization coils between 1 January 2014 and 31 December 2018. Available clinical follow-up for these patients was 1127 ± 719 days. Indications for use, adverse events, and procedural success rates were identified via automated extraction of structured data along with review of available unstructured data. The overall technical success rate was 96.7%, and the safety events rate was 5.3% with 18 major adverse events in 17 patients. The calculated technical success and safety rates met pre-established performance goals (≥ 85% for technical success and ≤ 12% for safety), highlighting the relevance of this surveillance method.ConclusionsGenerating RWE from RWD requires careful planning and execution. The process described herein provided valuable longitudinal data for PMS of real-world device safety and performance. This cost-effective approach can be translated to other medical devices and similar RWD database systems.

Physician–patient racial concordance and newborn mortality

The racial gap in infant mortality is a pressing public-health concern, and [B. N. Greenwood et al., Proc. Natl. Acad. Sci. U.S.A. 117, 21194-21200 (2020), 10.1073/pnas.1913405117] suggest that Black newborns are more likely to survive if cared for by Black physicians after birth, even in models that control for numerous variables, including hospital and physician fixed effects, and the 65 most common comorbidities affecting newborns (as described by International Classification of Disease codes). We acquired the data used in the study, covering Florida hospital discharges from 1992 through the third quarter of 2015, to replicate and extend the analysis. We find that the magnitude of the concordance effect is substantially reduced after controlling for diagnoses indicating very low birth weight (<1,500 g), which are a strong predictor of neonatal mortality but not among the 65 most common comorbidities. In fact, the estimated effect is near zero and statistically insignificant in the expanded specifications that control for very low birth weight and include hospital and physician fixed effects.

Repeated CT scans on 12,984 Asians to diagnose lung cancer once it is suspected.

In Taiwan, lung cancer remains the leading cause of cancer-related fatalities, resulting in substantial healthcare expenses. This research aims to evaluate both the frequency and the costs of low-dose computed tomography (LDCT) in individuals suspected of having lung cancer until their diagnosis of cancer. LDCT screening was not conducted on a population-wide scale, and asymptomatic participants had to cover the expenses for the screening personally or reimburse from other sources. If the screening results were positive or suspicious, National Health Insurance (NHI) could be utilized for subsequent follow-up examinations. This cohort study utilized the NHI Database and focused on individuals with suspected cases of lung cancer identified between 2010 and 2014.A total of 17,572 suspected new lung cancer cases were initially identified and assigned to the relevant International Classification of Diseases codes. Individuals with suspected lung cancer received a diagnosis following an average follow-up period of 2.24 (95%CI, 2.11-2.37) years, and required the use of 2.36 (95%CI, 2.20-2.51) repeated CT scans. The NHI expenditures incurred by the use of CT scans for monitoring suspected lung cancer cases were relatively modest.

Association between MASLD and increased risk of serious bacterial infections requiring hospital admission: A meta-analysis.

Previous studies have reported an association between metabolic dysfunction-associated steatotic liver disease (MASLD) and the risk of serious bacterial infections. However, the magnitude of the risk and whether this risk varies with the severity of MASLD remains uncertain. We performed a meta-analysis of observational studies to quantify the association between MASLD and serious bacterial infections requiring hospital admission. We systematically searched PubMed, Scopus, Web of Science and Embase from database inception to 1 April 2024, using predefined keywords to identify studies examining the risk of serious bacterial infections among individuals with and without MASLD. MASLD was diagnosed using liver biopsy, imaging or International Classification of Diseases codes. Meta-analysis was performed using random-effects modelling. We identified six cross-sectional and two prospective cohort studies with aggregate data on ~26.6 million individuals. MASLD was significantly associated with higher odds of serious bacterial infections (pooled random-effects odds ratio 1.93, 95% confidence interval [CI] 1.44-2.58; I2 = 93%). Meta-analysis of prospective cohort studies showed that MAFLD was associated with an increased risk of developing serious bacterial infections (pooled random-effects hazard ratio 1.80, 95% CI 1.62-2.0; I2 = 89%). This risk further increased across the severity of MASLD, especially the severity of fibrosis (pooled random-effects hazard ratio 2.42, 95% CI 1.89-2.29; I2 = 92%). These results remained significant after adjusting for age, sex, obesity, diabetes and other potential confounders. Sensitivity analyses did not modify these findings. The funnel plot did not reveal any significant publication bias. This meta-analysis shows a significant association between MASLD and an increased risk of serious bacterial infections requiring hospital admission.

Minimum National Prevalence of Diagnosed Atrial Fibrillation Inferred From California Acute Care Facilities

BackgroundPrevalence estimates of atrial fibrillation (AF) from large populations have not been updated for >2 decades. Using data from 1996 to 1997, a previous study projected that there would be 3.3 million adults with AF in the United States in 2020. ObjectivesThe purpose of this study was to determine the contemporary age-, sex-, and race-standardized prevalence and the number of adults with diagnosed AF in the United States. MethodsWe merged California’s state-wide health care databases to assemble a cohort of adults aged ≥20 years who received hospital-based care in California from 2005 to 2019. International Classification of Diseases codes were used to identify AF and other comorbidities. After accounting for deaths, we utilized the U.S. Census to calculate the national age-, sex-, and race-standardized estimates of diagnosed AF. ResultsOf 29,250,310 patients (mean age 50.6 ± 19.8 years, 53.8% women, 50.1% White), 2,003,867 (6.8%) had an AF diagnosis. The proportion of patients with diagnosed AF increased from 4.49% in 2005 to 2009 to 6.82% in 2015 to 2019. Over time, AF patients became relatively younger, were less likely to be female or White, and were more likely to have hypertension and diabetes. Standardizing based on age-, sex-, race-, and ethnicity-based proportions to the U.S. population, we estimate that the current national prevalence of diagnosed AF is at least 10.55 million (95% CI: 10.48-10.62 million), comprising 4.48% (95% CI: 4.47%-4.49%) of the adult population. ConclusionsThe prevalence of diagnosed AF in the United States is higher than previously estimated. More efficient prevention and treatment strategies are needed to curb the burden of AF in the United States.

Improving Detection of Hospital Adverse Events Using Machine Learning on Real-World Narrative EMR Data

ObjectiveAdministrative data often underrepresents hospital adverse events (AEs) due to limitations in International Classification of Diseases (ICD) coding. By leveraging electronic medical records (EMRs), we aim to mitigate these discrepancies and enhance the precision of healthcare surveillance and performance evaluations. To this end, we have developed a machine learning (ML)-based approach that utilizes EMR text data to detect common AEs. MethodsWe sampled adult admissions from four Calgary hospitals (2017 - 2022). Registered nurses assessed charts for 17 AEs, and the results were used as reference standard. We compared two AE detection methods: the standard ICD-based method following Canadian guidelines, and our ML algorithm applied to EMR narratives. Sensitivity, positive predictive value (PPV), negative predictive value (NPV), and specificity for both methods were calculated and compared against the reference standard. ResultsWe analyzed 9,566 patients, of whom 1,506 were identified with AEs. Of the 17 AEs, the sensitivity in ICD-coded data ranged from 0-37%, and in EMRs, it was between 75-100%. Both showed low PPV (0-50% ICD vs.1-34% EMR). ICD data had high specificity ranging from 99-100% and NPV (99%-100%), while EMRs had specificity between 68-94% and an NPV of 100%. Conclusion and ImplicationsML significantly enhances sensitivity for AE detection compared to ICD-10-CA coding, despite both methods experiencing low PPV due to imbalances in EMR data. This marked improvement in sensitivity highlights ML's potential to transform AE surveillance and reporting, promising significant advancements in patient safety and healthcare quality by enabling more accurate and comprehensive identification of AEs.