Chinese Electronic Medical Records Research Articles

Medical knowledge embedding based on recursive neural network for multi-disease diagnosis.

The representation of knowledge based on first-order logic captures the richness of natural language and supports multiple probabilistic inference models. Although symbolic representation enables quantitative reasoning with statistical probability, it is difficult to utilize with machine learning models as they perform numerical operations. In contrast, knowledge embedding (i.e., high-dimensional and continuous vectors) is a feasible approach to complex reasoning that can not only retain the semantic information of knowledge, but also establish the quantifiable relationship among embeddings. In this paper, we propose a recursive neural knowledge network (RNKN), which combines medical knowledge based on first-order logic with a recursive neural network for multi-disease diagnosis. After the RNKN is efficiently trained using manually annotated Chinese Electronic Medical Records (CEMRs), diagnosis-oriented knowledge embeddings and weight matrixes are learned. The experimental results confirm that the diagnostic accuracy of the RNKN is superior to those of four machine learning models, four classical neural networks and Markov logic network. The results also demonstrate that the more explicit the evidence extracted from CEMRs, the better the performance. The RNKN gradually reveals the interpretation of knowledge embeddings as the number of training epochs increases.

Named Entity Recognition in Chinese Electronic Medical Records Based on the Model of Bidirectional Long Short-Term Memory with a Conditional Random Field Layer.

Named entity recognition in electronic medical records is of great significance to the construction of medical knowledge maps. This paper proposes a model of bidirectional Long Short-Term Memory with a conditional random field layer(BiLSTM-CRF). In terms of simultaneously identifying 5 types of clinical entities from CCKS2018 Chinese EHRs corpus, the BiLSTM-CRF model finally achieved better performance than the baseline CRF model (F-score of 84.23% vs 82.49%).

Exploiting the concept level feature for enhanced name entity recognition in Chinese EMRs

The accumulation and explosive growth of the electronic medical records (EMRs) make the name entity recognition (NER) technologies become critical for the meaningful use of EMR data and then the practice of evidence-based medicine. The dominate NER approaches use the distributed representation of the words and characters to build deep learning-based NER models. However, for the task of biomedical named entity recognition, there are a large amount of complicated medical terminologies that are composed of multiple words. Splitting these terminologies to learn the word and character embeddings might cause semantic ambiguities. In this paper, we treat each medical terminology as a concept and propose a concept-enhanced named entity recognition model (CNER), where the features from three different granularities (i.e., concept, word, and character) are combined together for bio-NER. The extensive experiments are conducted on two real-world corpora: fully labeled corpus and partially labeled corpus. CNER achieves the highest F1 score (fully labeled corpus: precision = 88.23, recall = 88.29, and F1 = 88.26; partially labeled corpus: precision = 87.03, recall = 88.19, and F1 = 87.61) by outperforming the baseline CW-BLSTM-CRF approach for 0.58% and 1.15% respectively, which demonstrates the effectiveness of the proposed approach.

A hybrid approach for named entity recognition in Chinese electronic medical record

BackgroundWith the rapid spread of electronic medical records and the arrival of medical big data era, the application of natural language processing technology in biomedicine has become a hot research topic.MethodsIn this paper, firstly, BiLSTM-CRF model is applied to medical named entity recognition on Chinese electronic medical record. According to the characteristics of Chinese electronic medical records, obtain the low-dimensional word vector of each word in units of sentences. And then input the word vector to BiLSTM to realize automatic extraction of sentence features. And then CRF performs sentence-level word tagging. Secondly, attention mechanism is added between the BiLSTM and the CRF to construct Attention-BiLSTM-CRF model, which can leverage document-level information to alleviate tagging inconsistency. In addition, this paper proposes an entity auto-correct algorithm to rectify entities according to historical entity information. At last, a drug dictionary and post-processing rules are well-built to rectify entities, to further improve performance.ResultsThe final F1 scores of the BiLSTM-CRF and Attention-BiLSTM-CRF model on given test dataset are 90.15 and 90.82% respectively, both of which are higher than 89.26%, which is the best F1 score on the test dataset except ours.ConclusionOur approach can be used to recognize medical named entity on Chinese electronic medical records and achieves the state-of-the-art performance on the given test dataset.

Constructing a Chinese electronic medical record corpus for named entity recognition on resident admit notes

BackgroundElectronic Medical Records(EMRs) contain much medical information about patients. Medical named entity extracting from EMRs can provide value information to support doctors’ decision making. The research on information extraction of Chinese Electronic Medical Records is still behind that has done in English.MethodsThis paper proposed a practical annotation scheme for medical entity extraction on Resident Admit Notes (RANs), and a model which can automatic extract medical entity. Nine types of clinical entities, four types of clinical relationships were defined in our annotation scheme. An end-to-end deep neural network with convolution neural network and long-short term memory units was applied in our model for the medical named entity recognition(NER).ResultWe annotated RANs in three rounds. The overall F-score of annotation consistency was up to 97.73%. And our NER model on the 255 annotated RANs achieved the best F-score of 91.08%.ConclusionThe annotation scheme and the model for NER in this paper are effective to extract medical named entity from RANs and provide the basis for fully excavating the patient’s information.

A deep learning model incorporating part of speech and self-matching attention for named entity recognition of Chinese electronic medical records

BackgroundThe Named Entity Recognition (NER) task as a key step in the extraction of health information, has encountered many challenges in Chinese Electronic Medical Records (EMRs). Firstly, the casual use of Chinese abbreviations and doctors’ personal style may result in multiple expressions of the same entity, and we lack a common Chinese medical dictionary to perform accurate entity extraction. Secondly, the electronic medical record contains entities from a variety of categories of entities, and the length of those entities in different categories varies greatly, which increases the difficult in the extraction for the Chinese NER. Therefore, the entity boundary detection becomes the key to perform accurate entity extraction of Chinese EMRs, and we need to develop a model that supports multiple length entity recognition without relying on any medical dictionary.MethodsIn this study, we incorporate part-of-speech (POS) information into the deep learning model to improve the accuracy of Chinese entity boundary detection. In order to avoid the wrongly POS tagging of long entities, we proposed a method called reduced POS tagging that reserves the tags of general words but not of the seemingly medical entities. The model proposed in this paper, named SM-LSTM-CRF, consists of three layers: self-matching attention layer – calculating the relevance of each character to the entire sentence; LSTM (Long Short-Term Memory) layer – capturing the context feature of each character; CRF (Conditional Random Field) layer – labeling characters based on their features and transfer rules.ResultsThe experimental results at a Chinese EMRs dataset show that the F1 value of SM-LSTM-CRF is increased by 2.59% compared to that of the LSTM-CRF. After adding POS feature in the model, we get an improvement of about 7.74% at F1. The reduced POS tagging reduces the false tagging on long entities, thus increases the F1 value by 2.42% and achieves an F1 score of 80.07%.ConclusionsThe POS feature marked by the reduced POS tagging together with self-matching attention mechanism puts a stranglehold on entity boundaries and has a good performance in the recognition of clinical entities.

Attention-based deep residual learning network for entity relation extraction in Chinese EMRs

BackgroundElectronic medical records (EMRs) contain a variety of valuable medical concepts and relations. The ability to recognize relations between medical concepts described in EMRs enables the automatic processing of clinical texts, resulting in an improved quality of health-related data analysis. Driven by the 2010 i2b2/VA Challenge Evaluation, the relation recognition problem in EMRs has been studied by many researchers to address this important aspect of EMR information extraction.MethodsThis paper proposes an Attention-Based Deep Residual Network (ResNet) model to recognize medical concept relations in Chinese EMRs.ResultsOur model achieves F1-score of 77.80% on the manually annotated Chinese EMRs corpus and outperforms the state-of-the-art approaches.ConclusionThe residual network-based model can reduce the negative impact of corpus noise to parameter learning, and the combination of character position attention mechanism will enhance the identification features of different type of entities.

Intelligent diagnosis with Chinese electronic medical records based on convolutional neural networks

BackgroundBenefiting from big data, powerful computation and new algorithmic techniques, we have been witnessing the renaissance of deep learning, particularly the combination of natural language processing (NLP) and deep neural networks. The advent of electronic medical records (EMRs) has not only changed the format of medical records but also helped users to obtain information faster. However, there are many challenges regarding researching directly using Chinese EMRs, such as low quality, huge quantity, imbalance, semi-structure and non-structure, particularly the high density of the Chinese language compared with English. Therefore, effective word segmentation, word representation and model architecture are the core technologies in the literature on Chinese EMRs.ResultsIn this paper, we propose a deep learning framework to study intelligent diagnosis using Chinese EMR data, which incorporates a convolutional neural network (CNN) into an EMR classification application. The novelty of this paper is reflected in the following: (1) We construct a pediatric medical dictionary based on Chinese EMRs. (2) Word2vec adopted in word embedding is used to achieve the semantic description of the content of Chinese EMRs. (3) A fine-tuning CNN model is constructed to feed the pediatric diagnosis with Chinese EMR data. Our results on real-world pediatric Chinese EMRs demonstrate that the average accuracy and F1-score of the CNN models are up to 81%, which indicates the effectiveness of the CNN model for the classification of EMRs. Particularly, a fine-tuning one-layer CNN performs best among all CNNs, recurrent neural network (RNN) (long short-term memory, gated recurrent unit) and CNN-RNN models, and the average accuracy and F1-score are both up to 83%.ConclusionThe CNN framework that includes word segmentation, word embedding and model training can serve as an intelligent auxiliary diagnosis tool for pediatricians. Particularly, a fine-tuning one-layer CNN performs well, which indicates that word order does not appear to have a useful effect on our Chinese EMRs.

Extraction of risk factors for cardiovascular diseases from Chinese electronic medical records

Background and objectiveEarly prevention of cardiovascular diseases (CVDs) can effectively prevent later loss of health, and the detection of CVDs risk factors is a simple method to achieve early prevention. Personal health records play a prominent role in the field of health information extraction because of their factuality and reliability. This present study describes how to extract risk factors for CVDs from Chinese electronic medical records (CEMRs). MethodsThe extraction process involves two tasks: (a) CVDs risk factor recognition and (b) risk factor time and assertion classification. We considered risk factor recognition as a named entity recognition (NER) task and time and assertion classification as a textual classification task. An information extraction pipeline system consisting of NER and textual classification modules with machine learning models was developed. In the risk factor recognition module, bidirectional long short term memory (BLSTM) with extra risk factor textual feature input was built, as well, convolutional neural networks (CNNs) with risk factor type and section label input and support vector machine (SVM) were built for time and assertion classification. ResultsWe have achieved the best performance of risk factor recognition with F1 value of 0.9609, time and assertion classification with F1 of 0.9812 and 0.9612, respectively. The experimental results showed that our system achieved a high performance and can extract risk factors from CEMRs efficiently. ConclusionsThe proposed system is the first system for CVDs risk factors extraction from CEMRs and shows competition to risk factor extraction systems that developed on English EMRs. Further, its good performance should have a strong influence on CVDs prevention.

Using natural language processing to extract clinically useful information from Chinese electronic medical records

AimsTo develop a natural language processing (NLP)-based algorithm for extracting clinically useful information for patients with hepatocellular carcinoma (HCC) from Chinese electronic medical records (EMRs) and use these data for the assessment of HCC staging. Materials and MethodsClinical documents, including operation notes, radiology and pathology reports, of 92 HCC patients were collected from Chinese EMRs. We randomly grouped these patients into training (n = 60) and testing (n = 32) datasets. Rule-based and hybrid methods for extracting information were developed using the training set of manually-annotated operation notes. The method with better performance was used to process other documents. The performance of the algorithm was assessed via calculating the precision, recall and F-score for exact-boundary and partial-boundary matching strategies. The utility of clinically useful information for the HCC staging was assessed in comparison with that manually reviewed. ResultsFor operation notes, the rule-based and hybrid methods had a precision, recall and F-score ≥80% when the exact-boundary and partial-boundary matching strategies were applied to the testing dataset. By using the rule-based method (which has better performance than the hybrid method), three other types of documents also obtained good performance. When the extracted clinically useful information was applied for the HCC staging, the concordance rate with the manual review was 75%. ConclusionA NLP system was developed for clinical information extraction and HCC staging based on EMRs, and the results indicate that Chinese NLP has potential utility in clinical research.

Medical assertion classification in Chinese EMRs using attention enhanced neural network.

Electronic medical records (EMRs), such as hospital discharge summaries, contain a wealth of information only expressed in natural language. Automated methods for extracting information from these records must be able to recognize medical concepts in text and their semantic context. A contextual property critical to reason on information from EMRs is the doctor's belief status or assertion of the patient's medical problem. Research on the medical assertion classification (MAC) can establish the foundation for various health data analyses and clinical applications. However, previous MAC studies are mainly based on traditional machine learning methods which mostly require manually constructed features and the original unlabeled data cannot be easily and effectively applied to classification or classification tasks. Furthermore, external medical knowledge such as various medical dictionary bases, which provides rich explain and definition information about medical entity, is rarely utilized in existing neural network models of medical information extraction. In this study, we propose a deep neural network architecture enhanced by medical knowledge attention layer through combining GRU neural network with CNN model to classify the assertion type of medical problem such as disease and symptom in Chinese EMRs. The attention layer in the model is applied to integrate entity representations learned from medical dictionary bases as query for encoding. Experimental results on own manually annotated corpus indicate our approach achieves better performance compared to existing methods.

A Hybrid Method to Extract Clinical Information From Chinese Electronic Medical Records

Narrative reports in medical records contain abundant clinical information that may be converted into structured data for managing patient information and predicting trends in diseases. Though various rule-based and machine-learning methods are available in electronic medical records (EMRs), a few works have explored the hybrid methods in extracting information from the Chinese EMRs. In this paper, we developed a novel hybrid approach which integrates the rules and bidirectional long short-term memory with a conditional random field layer (BiLSTM-CRF) model to extract clinical entities and attributes. A corpus of 1509 electronic notes (discharge summaries and operation notes) was annotated. Annotation from three clinicians was reconciled to form a gold standard dataset. The performance of our method was assessed by calculating the precision, recall, and F-measure for two boundary matching strategies. The experimental results demonstrate the effectiveness of our method in clinical information extraction from the Chinese EMRs.

A Text Structuring Method for Chinese Medical Text Based on Temporal Information.

Chinese Electronic Medical Records (EMRs) contains a large number of complex medical free text which includes a variety of information, such as temporal information, patients’ symptoms and laboratory data. However, as an important knowledge base, these unstructured text data in EMR are hard to process directly by computer to support further medical research. This paper proposes a novel text structuring method to extract knowledge from EMR texts and reorganize them in chronological order according to the temporal information in the text. By implementing some entropy-based algorithms as contrast, experiments evaluate the performance of the proposed method, which indicates the new method can significantly reduce the complexity of EMR text. This work is significant in structuring the EMR free text into temporal-structured data for further medical analysis.

WCP-RNN: a novel RNN-based approach for Bio-NER in Chinese EMRs

Deep learning has achieved remarkable success in a wide range of domains. However, it has not been comprehensively evaluated as a solution for the task of Chinese biomedical named entity recognition (Bio-NER). The traditional deep-learning approach for the Bio-NER task is usually based on the structure of recurrent neural networks (RNN) and only takes word embeddings into consideration, ignoring the value of character-level embeddings to encode the morphological and shape information. We propose an RNN-based approach, WCP-RNN, for the Chinese Bio-NER problem. Our method combines word embeddings and character embeddings to capture orthographic and lexicosemantic features. In addition, POS tags are involved as a priori word information to improve the final performance. The experimental results show our proposed approach outperforms the baseline method; the highest F-scores for subject and lesion detection tasks reach 90.36 and 90.48% with an increase of 3.10 and 2.60% compared with the baseline methods, respectively.

Max-margin weight learning for medical knowledge network

Background and objectiveThe application of medical knowledge strongly affects the performance of intelligent diagnosis, and method of learning the weights of medical knowledge plays a substantial role in probabilistic graphical models (PGMs). The purpose of this study is to investigate a discriminative weight-learning method based on a medical knowledge network (MKN). MethodsWe propose a training model called the maximum margin medical knowledge network (M3KN), which is strictly derived for calculating the weight of medical knowledge. Using the definition of a reasonable margin, the weight learning can be transformed into a margin optimization problem. To solve the optimization problem, we adopt a sequential minimal optimization (SMO) algorithm and the clique property of a Markov network. Ultimately, M3KN not only incorporates the inference ability of PGMs but also deals with high-dimensional logic knowledge. ResultsThe experimental results indicate that M3KN obtains a higher F-measure score than the maximum likelihood learning algorithm of MKN for both Chinese Electronic Medical Records (CEMRs) and Blood Examination Records (BERs). Furthermore, the proposed approach is obviously superior to some classical machine learning algorithms for medical diagnosis. To adequately manifest the importance of domain knowledge, we numerically verify that the diagnostic accuracy of M3KN is gradually improved as the number of learned CEMRs increase, which contain important medical knowledge. ConclusionsOur experimental results show that the proposed method performs reliably for learning the weights of medical knowledge. M3KN outperforms other existing methods by achieving an F-measure of 0.731 for CEMRs and 0.4538 for BERs. This further illustrates that M3KN can facilitate the investigations of intelligent healthcare.

Learning and inference in knowledge-based probabilistic model for medical diagnosis

Based on a weighted knowledge graph to represent first-order knowledge and combining it with a probabilistic model, we propose a methodology for creating a medical knowledge network (MKN) in medical diagnosis. When a set of evidence is activated for a specific patient, we can generate a ground medical knowledge network that is composed of evidence nodes and potential disease nodes. By incorporating a Boltzmann machine into the potential function of a Markov network, we investigated the joint probability distribution of the MKN. To consider numerical evidence, a multivariate inference model is presented that uses conditional probability. In addition, the weights for the knowledge graph are efficiently learned from manually annotated Chinese Electronic Medical Records (CEMRs) and Blood Examination Records (BERs). In our experiments, we found numerically that an improved expression of evidence variables is necessary for medical diagnosis. Our experimental results comparing a Markov logic network and six kinds of classic machine learning algorithms on the actual CEMR database and BER database indicate that our method holds promise and that MKN can facilitate studies of intelligent diagnosis.

Developing a cardiovascular disease risk factor annotated corpus of Chinese electronic medical records

BackgroundCardiovascular disease (CVD) has become the leading cause of death in China, and most of the cases can be prevented by controlling risk factors. The goal of this study was to build a corpus of CVD risk factor annotations based on Chinese electronic medical records (CEMRs). This corpus is intended to be used to develop a risk factor information extraction system that, in turn, can be applied as a foundation for the further study of the progress of risk factors and CVD.ResultsWe designed a light annotation task to capture CVD risk factors with indicators, temporal attributes and assertions that were explicitly or implicitly displayed in the records. The task included: 1) preparing data; 2) creating guidelines for capturing annotations (these were created with the help of clinicians); 3) proposing an annotation method including building the guidelines draft, training the annotators and updating the guidelines, and corpus construction. Meanwhile, we proposed some creative annotation guidelines: (1) the under-threshold medical examination values were annotated for our purpose of studying the progress of risk factors and CVD; (2) possible and negative risk factors were concerned for the same reason, and we created assertions for annotations; (3) we added four temporal attributes to CVD risk factors in CEMRs for constructing long term variations. Then, a risk factor annotated corpus based on de-identified discharge summaries and progress notes from 600 patients was developed. Built with the help of clinicians, this corpus has an inter-annotator agreement (IAA) F1-measure of 0.968, indicating a high reliability.ConclusionTo the best of our knowledge, this is the first annotated corpus concerning CVD risk factors in CEMRs and the guidelines for capturing CVD risk factor annotations from CEMRs were proposed. The obtained document-level annotations can be applied in future studies to monitor risk factors and CVD over the long term.

Developing and Validating Electronic Medical Record Based Case Definitions for Liver Diseases and Comorbidities

ABSTRACTObjectiveChina has collected high volume of electronic health record (EMR) data. The rich information in EMR data could be used for health services research. The first challenge is developing methods for extracting study variables. Our study aimed to develop and validate data extraction methods for defining clinical conditions.
 ApproachThe EMRs were from Beijing You-An Hospital, a leading liver diseases specialized teaching hospital affiliated with Capital Medical University in China. We developed EMR based case definitions for extracting common liver diseases and comorbidities in Charlson and Elixhauser comorbidity algorithms. We developed the EMR case definitions based on the fundamental EMR structure and clinical expertise. To determine validity of the EMR case definitions, we calculated the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) for each case definition based on the “gold standard”: randomly selected 450 charts review conducted by two hepatologists. The agreement between the two reviewers was assessed by kappa value.
 ResultsIn total, 69,864 EMRs for adult patients with liver disease admitted between 2010 and 2015 were included in this study. Among these patients, we identified 13,763 (19.7%) PLC, 18,654 (26.7%) Hepatitis B Virus (HBV), 4681 (6.7%) Hepatitis C Virus (HCV), 18,933 (27.1%) cirrhosis, 7,964 (11.4%) diabetes, 8,523 (12.2%) hypertension, 3,283 (4.7%) Fluid and Electrolyte Disorder (FED), 3,563 (5.1%) renal disease and 1,258 (1.8%) Cerebrovascular Disease (CEVD). Between the two reviewers, the Kappa values fell between 0.65 and 1.00. Of the 450 EMRs reviewed, the two reviewers identified 95 (21.1%) PLC, 197 (43.8%) HBV, 40 (8.9%) HCV, 146 (32.4%) cirrhosis, 39 (8.7%) diabetes, 45 (10.0%) hypertension, 25 (5.6%) FED, 26 (5.8%) renal disease and 8 (1.8%) CEVD. Compared to chart review, the sensitivity, specificity, PPV and NPV of the algorithms for above conditions are respectively: (PLC) 100%, 99.44%, 97.94%, 100%; (HBV) 61.93%, 99.21%, 98.39%, 76.99%; (HCV) 72.50%, 99.51%, 93.55%, 97.37%; (cirrhosis) 78.77%, 98.36%, 95.83%, 90.61%; (diabetes) 97.44%, 98.05%, 82.61%, 99.75%; (hypertension) 95.00%, 99.72%, 97.44%, 99.45%; (FED) 71.43%, 100.00%, 100.00%, 99.09%; (renal disease) 96.15% 100.00% 100.00% 99.76%; (CEVD) 100.00%, 99.77%, 88.89%, 100.00%.
 ConclusionOur EMR case definitions of above conditions had high validity and could be applied to extract clinical variables including major liver diseases and comorbidities from Chinese EMR. This extracting method could be modified slightly for extracting other medical conditions from Chinese EMRs.

Exploring the Impact of Health Insurance on Health Care Utilization and Outcome Using Electronic Medical Record Data

ABSTRACTObjectiveWith tremendous potential for research and policy use, the development of Electronic Medical Record (EMR) is unprecedentedly growing in China. The rich clinical and financial data in the Chinese EMR provides us a unique chance to examine the impacts of health insurance on health care utilization and outcomes, controlling for patient’s disease severity. ApproachesOur study population included patients with cirrhosis or primary liver cancer (PLC), from a large teaching hospital in Beijing. The comorbidity and disease severity variables were defined using EMR automated extraction methods that were validated in previous study. Health insurance was measured by actual reimbursement ratio (RR), which better captures patients’ actual financial burden than type of health insurance. Generalized linear regression model was used to analyze the impacts of health insurance coverage on total hospital expense, ratio of medication cost to total expense, and number of major procedures (i.e., transcatheter arterial chemoembolization, TACE) for cirrhosis. Logistic regression was used to assess the impact of health insurance on hospital mortality and the rate of TACE. We employed a wide range of risk factors in our models to adjust for disease severity and comorbidities, including Charlson comorbidities, MELD-Na score, and etiological factors of liver diseases.ResultsIn total, 5,465 cirrhosis patients and 3,357 PLC patients were included in the study. Among the PLC patients we identified 534 patients underwent TACE. After adjusted for comorbidities, disease severity and other confounders, RR was found to be associated with hospital mortality with odds ratio 3.2 in cirrhosis patients and 6.0 in PLC patients. Higher RR was correlated to lower total hospital cost (logarithm transferred coefficient -0.08 in cirrhosis patients and -0.15 in PLC patients) but related to higher ratio of medication cost (logarithm transferred coefficient 0.09 in both cirrhosis and PLC patients). Additionally, higher RR was associated with higher rate (odds ratio 1.6 in PLC patients) and also more times of TACE (logarithm transferred coefficient 0.31 in TACE patients). The results were consistent between cirrhosis patients and PLC patients.ConclusionThis study provided evidences that physicians’ behavior was influenced by health insurance. Patients with more generous health insurance coverage (higher RR) were found to have relatively lower total hospital cost but higher ratio of medication cost, higher rate and more times of TACE, and were more prone to die in hospital. These are evidences for physician’s gaming reacting to the economic incentives of the payment systems in China.

Comparison of Risk Adjustment Methods in Patients with Liver Disease Using Electronic Medical Record

ABSTRACTObjectiveRisk adjustment methods are widely used to compare quality of care or predict health outcome, but the optimal approach is unclear for liver disease. This study is to compare the performance of common risk adjustment methods for predicting in-hospital mortality in patients with liver disease using Electronic Medical Record (EMR).
 ApproachThe EMR data was derived from Beijing YouAn Hospital between 2010 and 2015. 85,526 EMRs were included. Previously developed and validated automated EMR case definitions were applied to define the conditions including primary liver cancer, cirrhosis and other conditions included in Charlson, Elixhauser comorbidity algorithms, Child-Turcotte-Pugh (CTP) score and Model for End-Stage Liver Disease (MELD). Logistic regression was conducted and C-statistic was obtained to compare the performance of the different methods for predicting in-hospital mortality. To eliminate the effect of the model complexity on model performance, we compared Akaike Information Criterion (AIC) of different methods (smaller AIC is better).
 ResultIn total, we included three liver diseases cohort: 7,178 Primary Liver Cancer (PLC) patients, 11,121 cirrhosis patients and 7,298 cirrhosis without PLC patient. For PLC cohort, C-statistics of these compared indexes ranged from 0.72 to 0.84; AIC was between 4312.3 and 5048.4. For cirrhosis cohort, C-statistics of these compared indexes ranged from 0.73 to 0.83; AIC was between 4952.1 and 5788.2. For cirrhosis without PLC cohort, C-statistics ranged from 0.73 to 0.84; AIC was between 2608.3 and 3240.5. It was consistent across the three cohorts that MELD + sodium (MELD_Na) score (a variant of MELD score) had the highest C-statistic and lowest AIC; CTP had the lowest C-statistic and highest AIC. Integrating Charlson Comorbidity to MELD_Na, C-statistic improved to 0.86 and AIC reduced.
 ConclusionAmong the compared risk adjustment methods, MELD_Na performed best for predicting in-hospital mortality among patients with PLC or cirrhosis using Chinese EMRs. Adding clinical information to comorbidity algorithms improved the performance of the model.