Accurate Identification of Patients with Cirrhosis and Its Complications in the Electronic Health Record.

BACKGROUND: Cirrhosis represents a significant health burden; administrative data provide an important tool for research studies. AIMS: We aimed to understand the validity of current ICD-10 codes compared to previously used ICD-9 codes to identify patients with cirrhosis and its complications. METHODS: We identified 1981 patients presenting to MUSC between 2013 and 2019 with a diagnosis of cirrhosis. To validate the sensitivity of ICD codes, we reviewed the medical records of 200 patients for each associated ICD 9 and 10 codes. Sensitivity, specificity, and positive predictive value for each ICD code (individually or when combined) were calculated and univariate binary logistic models, for cirrhosis and its complications, predicted probabilities were used to calculate C-statistics. RESULTS: Single ICD 9 and 10 codes were similarly insensitive for detection of cirrhosis, with sensitivity ranging from 5 to 94%. However, ICD-9 code combinations (when used as either/or) had high sensitivity and specificity for the detection of cirrhosis, with the combination of either 571.5 (or 456.21) or 571.2 codes having a C-statistic of 0.975. Combinations of ICD-10 codes were only slightly less sensitive and specific than ICD-9 codes for detection of cirrhosis (K76.6, or K70.31, plus K74.60 or K74.69, and K70.30 had a C-statistic of 0.927). CONCLUSIONS: ICD-9 and ICD-10 codes when used alone were inaccurate for identifying cirrhosis. ICD-10 and ICD-9 codes had similar performance characteristics. Combinations of ICD codes exhibited the greatest sensitivity and specificity for detection of cirrhosis, and thus should be used to accurately identify cirrhosis.

Artificial intelligence to differentiate asthma from COPD in medico-administrative databases.

INTRODUCTION: Discriminating asthma from chronic obstructive pulmonary disease (COPD) using medico-administrative databases is challenging but necessary for medico-economic analyses focusing on respiratory diseases. Artificial intelligence (AI) may improve dedicated algorithms. OBJECTIVES: To assess performance of different AI-based approaches to distinguish asthmatics from COPD patients in medico-administrative databases where the clinical diagnosis is absent. An "Asthma COPD Overlap" category was defined to further test whether AI can detect complexity. METHODS: This study included 178,962 patients treated by two "R03" treatment prescriptions at least from January 2016 to December 2018 and managed by either a general practitioner and/or a pulmonologist participating in a permanent longitudinal observatory of prescription in ambulatory medicine (LPD). Clinical diagnoses are available in this database and were used as gold standards to develop diagnostic rules. Three types of AI approaches were explored using data restricted to demographics and treatment dispensations: multinomial regression, gradient boosting and recurrent neural networks (RNN). The best performing model (based on metric properties) was then applied to estimate the size of asthma and COPD populations based on a database (LRx) of treatment dispensations between July, 2018 and June, 2019. RESULTS: The best models were obtained with the boosting approach and RNN, with an overall accuracy of 68%. Performance metrics were better for asthma than COPD. Based on LRx data, the extrapolated numbers of patients treated for asthma and COPD in France were 3.7 and 1.2 million, respectively. Asthma patients were younger than COPD patients (mean, 49.9 vs. 72.1 years); COPD occurred mostly in men (68%) compared to asthma (33%). CONCLUSION: AI can provide models with acceptable accuracy to distinguish between asthma, ACO and COPD in medico-administrative databases where the clinical diagnosis is absent. Deep learning and machine learning (RNN) had similar performances in this regard.

Incorporating natural language processing to improve classification of axial spondyloarthritis using electronic health records.

OBJECTIVES: To develop classification algorithms that accurately identify axial SpA (axSpA) patients in electronic health records, and compare the performance of algorithms incorporating free-text data against approaches using only International Classification of Diseases (ICD) codes. METHODS: An enriched cohort of 7853 eligible patients was created from electronic health records of two large hospitals using automated searches (⩾1 ICD codes combined with simple text searches). Key disease concepts from free-text data were extracted using NLP and combined with ICD codes to develop algorithms. We created both supervised regression-based algorithms-on a training set of 127 axSpA cases and 423 non-cases-and unsupervised algorithms to identify patients with high probability of having axSpA from the enriched cohort. Their performance was compared against classifications using ICD codes only. RESULTS: NLP extracted four disease concepts of high predictive value: ankylosing spondylitis, sacroiliitis, HLA-B27 and spondylitis. The unsupervised algorithm, incorporating both the NLP concept and ICD code for AS, identified the greatest number of patients. By setting the probability threshold to attain 80% positive predictive value, it identified 1509 axSpA patients (mean age 53 years, 71% male). Sensitivity was 0.78, specificity 0.94 and area under the curve 0.93. The two supervised algorithms performed similarly but identified fewer patients. All three outperformed traditional approaches using ICD codes alone (area under the curve 0.80-0.87). CONCLUSION: Algorithms incorporating free-text data can accurately identify axSpA patients in electronic health records. Large cohorts identified using these novel methods offer exciting opportunities for future clinical research.

Incidence of major depression diagnoses in the Canadian Armed Forces: longitudinal analysis of clinical and health administrative data.

PURPOSE: Major depression is a leading cause of morbidity in military populations. However, due to a lack of longitudinal data, little is known about the rate at which military personnel experience the onset of new episodes of major depression. We used a new source of clinical and administrative data to estimate the incidence of major depression diagnoses in Canadian Armed Forces (CAF) personnel, and to compare incidence rates between demographic and occupational factors. METHODS: We extracted all data recorded in the electronic medical records of CAF Regular Force personnel, at every primary care and mental health clinical encounter since 2016. Using a 12-month lookback period, we linked data over time, and identified all patients with incident diagnoses of major depression. We then linked clinical data to CAF administrative records, and estimated incidence rates. We used multivariate Poisson regression to compare adjusted incidence rates between demographic and occupational factors. RESULTS: From January to December 2017, CAF Regular Force personnel were diagnosed with major depression at a rate of 29.2 new cases per 1000 person-years at risk. Female sex, age 30 years and older, and non-officer ranks were associated with significantly higher incidence rates. CONCLUSIONS: We completed the largest study to date on diagnoses of major depression in the Canadian military, and have provided the first estimates of incidence rates in CAF personnel. Our results can inform future mental health resource allocation, and ongoing major depression prevention efforts within the Canadian Armed Forces and other military organizations.

Automatic ICD code assignment of Chinese clinical notes based on multilayer attention BiRNN.

International Classification of Diseases (ICD) code is an important label of electronic health record. The automatic ICD code assignment based on the narrative of clinical documents is an essential task which has drawn much attention recently. When Chinese clinical notes are the input corpus, the nature of Chinese brings some issues that need to be considered, such as the accuracy of word segmentation and the representation of single Chinese characters which contain semantics. Taking the lengthy text of patient notes and the representation of Chinese words into account, we present a multilayer attention bidirectional recurrent neural network (MA-BiRNN) model to implement the assignment of disease codes. A hierarchical approach is used to represent the feature of discharge summaries without manual feature engineering. The combination of character level embedding and word level embedding can improve the representation of words. Attention mechanism is introduced into bidirectional long short term memory networks, which helps to solve the performance dropping problem when plain recurrent neural networks encounter long text sequences. The experiment is carried out on a real-world dataset containing 7732 admission records in Chinese and 1177 unique ICD-10 labels. The proposed model achieves 0.639 and 0.766 in F1-score on full-level code and block-level code, respectively. It outperforms the baseline neural network models and achieves the lowest Hamming loss value. Ablation analysis indicates that the multilevel attention mechanism plays a decisive role in the system for dealing with Chinese clinical notes.

Algorithms to identify COPD in health systems with and without access to ICD coding: a systematic review.

BACKGROUND: Chronic obstructive pulmonary disease (COPD) causes significant morbidity and mortality worldwide. Estimation of incidence, prevalence and disease burden through routine insurance data is challenging because of under-diagnosis and under-treatment, particularly for early stage disease in health care systems where outpatient International Classification of Diseases (ICD) diagnoses are not collected. This poses the question of which criteria are commonly applied to identify COPD patients in claims datasets in the absence of ICD diagnoses, and which information can be used as a substitute. The aim of this systematic review is to summarize previously reported methodological approaches for the identification of COPD patients through routine data and to compile potential criteria for the identification of COPD patients if ICD codes are not available. METHODS: A systematic literature review was performed in Medline via PubMed and Google Scholar from January 2000 through October 2018, followed by a manual review of the included studies by at least two independent raters. Study characteristics and all identifying criteria used in the studies were systematically extracted from the publications, categorized, and compiled in evidence tables. RESULTS: In total, the systematic search yielded 151 publications. After title and abstract screening, 38 publications were included into the systematic assessment. In these studies, the most frequently used (22/38) criteria set to identify COPD patients included ICD codes, hospitalization, and ambulatory visits. Only four out of 38 studies used methods other than ICD coding. In a significant proportion of studies, the age range of the target population (33/38) and hospitalization (30/38) were provided. Ambulatory data were included in 24, physician claims in 22, and pharmaceutical data in 18 studies. Only five studies used spirometry, two used surgery and one used oxygen therapy. CONCLUSIONS: A variety of different criteria is used for the identification of COPD from routine data. The most promising criteria set in data environments where ambulatory diagnosis codes are lacking is the consideration of additional illness-related information with special attention to pharmacotherapy data. Further health services research should focus on the application of more systematic internal and/or external validation approaches.

[Prevalence of dementia in university hospitals : Analysis of ICD-10 coding in 5 university hospitals in Germany over a 2-year period]. / Prävalenzen von Demenz in Universitätskliniken : Analyse von ICD-10-Kodierungen in fünf Universitätskliniken in Deutschland über zwei Geschäftsjahre.

BACKGROUND: Demographic changes result in a higher prevalence of patients suffering from dementia in hospital. In Germany, epidemiological data of this target group are scarce and prevalence rates from university hospitals (UH) are not available. The prevalence rates and distribution were analyzed on the basis of ICD-10-GM (German modification) routine data METHOD: A secondary analysis on ICD-10-GM main and secondary diagnoses of dementia from 2014 and 2015 from 5 UH was performed. All patients admitted to hospital for at least 24 h and ≥18 years old (2014 n = 187,168; 2015 n = 189,040) were included. A descriptive analysis for the >69-year-old group was carried out (2014, n = 67,111; 2015; n = 67,824). RESULTS: The 1­year prevalence (2014/2015) for all 5 UH for patients ≥18 years old was 1.3%/1.4% and for the >69-year-old group, 3.3%/3.5%. The prevalence rates between the five UH varied: for patients ≥18 years the range was 0.44-2.16% (2014) and 0.44-2.77% (2015) and for >69-year-olds 1.16-5.52% (2014) and 1.16-7.06% (2015). Most cases were correlated with major diagnostic categories of traumatology, cardiology, gastroenterology and neurology. CONCLUSION: Analysis of ICD-10-GM routine data can provide an indication of the prevalence of dementia in UH. Results of the >69-year-olds varied greatly between participating UH. The reasons for this might be different healthcare tasks, especially with respect to geriatric patients; however, it is also possible that assessment procedures are not standardized and unreliable and therefore the coding is invalid. A standardized procedure for the identification of people suffering from dementia is necessary.

Methodological and Practical Challenges in Synthesizing Occupational Cancer Studies.

Studies examining occupational exposures and cancer risk frequently report mixed findings; it is thus imperative for researchers to synthesize study results and identify any potential sources that explain such variabilities in study findings. However, when synthesizing study results using meta-analytic techniques, researchers often encounter a number of practical and methodological challenges. These challenges include (1) an incomparability of effect size measures due to large variations in research methodology; (2) a violation of the independence assumption for meta-analysis; (3) a violation of the normality assumption of effect size measures; and (4) a variation in cancer definitions across studies and changes in coding standards over time. In this paper, we first demonstrate these challenges by providing examples from a real dataset collected for a large meta-analysis project that synthesizes cancer mortality and incidence rates among firefighters. We summarize how each of these challenges has been handled in our meta-analysis. We conclude this paper by providing practical guidelines for handling challenges when synthesizing study findings from occupational cancer literature.

The rising incidence of rotator cuff repairs.

BACKGROUND: Rotator cuff repairs (RCRs) have become increasingly common. Several studies have shown variation in the indications for this procedure. We chose to track the incidence of RCRs in New York State (NYS) from 1995 to 2009. We hypothesized that after the introduction of the Current Procedural Terminology (CPT) code 29827 for arthroscopic RCR, there would be a significant increase in the rate of RCRs performed in NYS. MATERIALS AND METHODS: The NYS Department of Health's Statewide Planning and Research Cooperative System (SPARCS) database was queried for reported RCRs between the years 1995 and 2009. Using the International Classification of Diseases, Ninth Revision, Clinical Modification procedural code 83.63 and CPT codes 23410, 23412, 23420, and 29827, we collected and analyzed data on RCR procedures. RESULTS: A total of 168,780 RCRs were performed in NYS from 1995 to 2009. In 1995, the population incidence of RCRs was 23.5 per 100,000. In comparison, in 2009, the population incidence was 83.1 per 100,000, an increase of 238% (P < .0001). The percentage of individuals aged between 45 and 65 years undergoing RCR increased from 53.0% to 64.2% during this same period. CONCLUSIONS: There has been a notable increase in the volume of RCRs performed in NYS. In addition, after the introduction of CPT code 29827 in 2003, the increase in the incidence of RCRs became significantly more pronounced. LEVEL OF EVIDENCE: Level III, cross-sectional design, epidemiology study.

Non-Fatal Opioid Overdose and Abuse Visits to a University Hospital Emergency Department in Puerto Rico, 2009-2018.

OBJECTIVE: The objective was to describe opioid-use trends (2009-2018) at a university hospital emergency department (ED) in metropolitan San Juan, Puerto Rico. METHODS: The ED database of the University of Puerto Rico - Dr. Federico Trilla Hospital provided the data for the study. RESULTS: Non-fatal opioid overdoses surged 7.5-fold, increasing from 12.1 (±2.5) per 100,000 ED encounters for 2009 through 2016 to 91.2 (±8.7) per 100,000 ED encounters for 2017 through 2018 (P < .0001). Starting in summer 2017, the surge reached its peak in October after two major hurricanes. The opioid-related ED cases comprised 15.8% from 2009 through 2016, increasing to 67% in 2017 through 2018. Prior to October 2015, multiple drugs were mentioned in 65% of the opioid-related cases, decreasing to 37% of the total cases, thereafter. Cocaine was reported in combination with opioids in 53% of all opioid-related cases from August 2009 through September 2015, decreasing to 21% from October 2015 through December 2018, cannabis in 15 % and 10%, respectively, and alcohol in 10% and 6%, respectively. Amphetamines were mentioned once in combination with opioids. The overall male:female ratio for all opioid-related cases was 6.3 (rate: 8.8). CONCLUSION: The data show an increase in opioid-toxicity cases in the area served by the above-named hospital beginning in mid-2017. Opioid-related cases overwhelmingly involved male patients. More work is needed to establish islandwide trends.

Automated ICD coding for primary diagnosis via clinically interpretable machine learning.

BACKGROUND: Computer-assisted clinical coding (CAC) based on automated coding algorithms has been expected to improve the International Classification of Disease, tenth version (ICD-10) coding quality and productivity, whereas studies oriented to primary diagnosis auto-coding are limited in the Chinese context. OBJECTIVE: This study aims at developing a machine learning (ML) model for automated primary diagnosis ICD-10 coding. METHODS: A total of 71,709 admissions in Fuwai hospital were included to carry out this study, corresponding to 168 primary diagnosis ICD-10 codes. Based on clinical implications, two feature engineering methods were used to process discharge diagnosis and procedure texts into sequential features and sequential grouping features respectively by which two kinds of models were built and compared. One baseline model using one-hot encoding features was considered. Light Gradient Boosting Machine (LightGBM) was adopted as the classifier, and grid search and cross-validation were used to select the optimal hyperparameters. SHapley Additive exPlanations (SHAP) values were applied to give the interpretability of models. RESULTS: Our best prediction model was developed based on sequential grouping features. It showed good performance in the test phase with accuracy and macro-averaged F1 (Macro-F1) of 95.2% and 88.3% respectively. The comparison of the models demonstrated the effectiveness of the sequential information and the grouping strategy in boosting model performance (P-value < 0.01). Subgroup analysis of the best model on each individual code manifested that 91.1% of the codes achieved the F1 over 70.0%. CONCLUSIONS: Our model has been demonstrated its effectiveness for automated primary diagnosis coding in the Chinese context and its results are interpretable. Hence, it has the potential to assist clinical coders to improve coding efficiency and quality in Chinese inpatient settings.

A Computable Phenotype for Autosomal Dominant Polycystic Kidney Disease.

Background: A computable phenotype is an algorithm used to identify a group of patients within an electronic medical record system. Developing a computable phenotype that can accurately identify patients with autosomal dominant polycystic kidney disease (ADPKD) will assist researchers in defining patients eligible to participate in clinical trials and other studies. Our objective was to assess the accuracy of a computable phenotype using International Classification of Diseases 9th and 10th revision (ICD-9/10) codes to identify patients with ADPKD. Methods: We reviewed four random samples of approximately 250 patients on the basis of ICD-9/10 codes from the EHR from the Kansas University Medical Center database: patients followed in nephrology clinics who had ICD-9/10 codes for ADPKD (Neph+), patients seen in nephrology clinics without ICD codes for ADPKD (Neph-), patients who were not followed in nephrology clinics with ICD codes for ADPKD (No Neph+), and patients not seen in nephrology clinics without ICD codes for ADPKD (No Neph-). We reviewed the charts and determined ADPKD status on the basis of internationally accepted diagnostic criteria for ADPKD. Results: The computable phenotype to identify patients with ADPKD who attended nephrology clinics has a sensitivity of 99% (95% confidence interval [95% CI], 96.4 to 99.7) and a specificity of 84% (95% CI, 79.5 to 88.1). For those who did not attend nephrology clinics, the sensitivity was 97% (95% CI, 93.3 to 99.0), and a specificity was 82% (95% CI, 77.4 to 86.1). Conclusion: A computable phenotype using the ICD-9/10 codes can correctly identify most patients with ADPKD, and can be utilized by researchers to screen health care records for cohorts of patients with ADPKD with acceptable accuracy.

Relationship of Atrial Fibrillation to Outcomes in Patients Hospitalized for Chronic Obstructive Pulmonary Disease Exacerbation.

INTRODUCTION: Chronic Obstructive Pulmonary Disease (COPD) is a major cause of hospitalization and is associated with an increased incidence of atrial fibrillation (AF). The impact of AF on in-hospital outcomes, including mortality, in patients hospitalized for COPD exacerbation is not well elucidated. METHODS: We used the National Inpatient Sample database to examine discharges with the primary diagnosis of COPD exacerbation and compared mortality, length of stay and costs in patients with AF compared to those without AF. The study adjusted the outcomes for known cardiovascular risk factors and confounders using logistic regression and propensity score matching analysis. RESULTS: Among 1,377,795 discharges with COPD exacerbation, 16.6% had AF. Patients with AF were older and had more comorbidities. Mortality was higher (2.4%) in the AF group than in the no AF group (1%), p <0.001. After adjustment to age, sex and confounders, AF remained an independent predictor for mortality, OR:1.44 (95% CI 133 - 1.56, p <0.001), prolonged length of stay, OR:1.63 (95% CI 1.57 - 1.69, p <0.001) and increased cost, OR: 1.45 (95% CI: 1.40 - 1.49, p <0.001). CONCLUSIONS: among patients with COPD exacerbation, AF was associated with increased mortality and higher resource utilization.

Welcome to the ICD-10 code for sarcopenia.

The new ICD-10-CM (M62.84) code for sarcopenia represents a major step forward in recognizing sarcopenia as a disease. This should lead to an increase in availability of diagnostic tools and the enthusiasm for pharmacological companies to develop drugs for sarcopenia.