Development of an Algorithm to Identify Cases of Nonalcoholic Steatohepatitis Cirrhosis in the Electronic Health Record

Administrative coding for non-alcoholic fatty liver disease is accurate in Swedish patients

Background and Aims: Epidemiological studies of non-alcoholic fatty liver disease (NAFLD) frequently use the International Classification of Disease (ICD) codes to identify patients. The validity of such ICD codes in a Swedish setting is unknown. Here, we aimed to validate the administrative code for NAFLD in Sweden.Methods: In total, 150 patients with an ICD-10 code for NAFLD (K76.0) from the Karolinska University Hospital between 1 January 2015 and 3 November 2021 were randomly selected. Patients were classified as true or false positives for NAFLD by medical chart review and the positive predictive value (PPV) for the ICD-10 code corresponding to NAFLD was calculated.Results: The PPV of the ICD-10 code for NAFLD was 0.82 (95% confidence interval [CI] = 0.76-0.89). After exclusion of patients with diagnostic coding for other liver diseases or alcohol abuse disorder (n = 14), the PPV was improved to 0.91 (95% CI 0.87-0.96). The PPV was higher in patients with coding for NAFLD in combination with obesity (0.95, 95%CI = 0.87-1.00) or type 2 diabetes (0.96, 95%CI = 0.89-1.00). However, in false-positive cases, a high alcohol consumption was common and such patients had somewhat higher Fibrosis-4 scores than true-positive patients (1.9 vs 1.3, p = 0.16)Conclusions: The ICD-10 code for NAFLD had a high PPV, that was further improved after exclusion of patients with coding for other liver diseases than NAFLD. This approach should be preferred when performing register-based studies to identify patients with NAFLD in Sweden. Still, residual alcohol-related liver disease might risk confound some findings seen in epidemiological studies which needs to be considered.

Electronic Health Record Burden Among Gastroenterology Providers Associated With Subspecialty and Training

INTRODUCTION

Use of the electronic health record (EHR) has become increasingly widespread. Higher EHR burden is associated with burnout, but this has not been specifically investigated among gastroenterology (GI) providers.

METHODS

We retrospectively collected measures of EHR use for outpatient GI providers during a 6-month period. We compared metrics across provider sex, subspecialty, and training (physicians vs nonphysician providers [NPPs]).

RESULTS

Data collected represented more than 16,000 appointments from 41 providers across the Division of Gastroenterology and Hepatology. Inflammatory bowel disease (IBD) and hepatology specialists spent more time per appointment in the EHR, clinical review, and outside regular hours compared with other subspecialists. NPPs spent more EHR time than physicians.

DISCUSSION

IBD and hepatology specialists and NPPs may have disproportionally high EHR burden. More work is needed to understand differences in provider workload to combat burnout.

Artificial intelligence in nonalcoholic fatty liver disease

Background

Nonalcoholic fatty liver disease (NAFLD) has led to serious health-related complications worldwide. NAFLD has wide pathological spectra, ranging from simple steatosis to hepatitis to cirrhosis and hepatocellular carcinoma. Artificial intelligence (AI), including machine learning and deep learning algorithms, has provided great advancement and accuracy in identifying, diagnosing, and managing patients with NAFLD and detecting squeal such as advanced fibrosis and risk factors for hepatocellular cancer. This review summarizes different AI algorithms and methods in the field of hepatology, focusing on NAFLD.

Methods

A search of PubMed, WILEY, and MEDLINE databases were taken as relevant publications for this review on the application of AI techniques in detecting NAFLD in suspected population

Results

Out of 495 articles searched in relevant databases, 49 articles were finally included and analyzed. NASH-Scope model accurately distinguished between NAFLD and non-NAFLD and between NAFLD without fibrosis and NASH with fibrosis. The logistic regression (LR) model had the highest accuracy, whereas the support vector machine (SVM) had the highest specificity and precision in diagnosing NAFLD. An extreme gradient boosting model had the highest performance in predicting non-alcoholic steatohepatitis (NASH). Electronic health record (EHR) database studies helped the diagnose NAFLD/NASH. Automated image analysis techniques predicted NAFLD severity. Deep learning radiomic elastography (DLRE) had perfect accuracy in diagnosing the cases of advanced fibrosis.

Conclusion

AI in NAFLD has streamlined specific patient identification and has eased assessment and management methods of patients with NAFLD.

Machine learning using longitudinal prescription and medical claims for the detection of non-alcoholic steatohepatitis (NASH)

Objectives

To develop and evaluate machine learning models to detect patients with suspected undiagnosed non-alcoholic steatohepatitis (NASH) for diagnostic screening and clinical management.

Methods

In this retrospective observational non-interventional study using administrative medical claims data from 1 463 089 patients, gradient-boosted decision trees were trained to detect patients with likely NASH from an at-risk patient population with a history of obesity, type 2 diabetes mellitus, metabolic disorder or non-alcoholic fatty liver (NAFL). Models were trained to detect likely NASH in all at-risk patients or in the subset without a prior NAFL diagnosis (at-risk non-NAFL patients). Models were trained and validated using retrospective medical claims data and assessed using area under precision recall curves and receiver operating characteristic curves (AUPRCs and AUROCs).

Results

The 6-month incidences of NASH in claims data were 1 per 1437 at-risk patients and 1 per 2127 at-risk non-NAFL patients. The model trained to detect NASH in all at-risk patients had an AUPRC of 0.0107 (95% CI 0.0104 to 0.0110) and an AUROC of 0.84. At 10% recall, model precision was 4.3%, which is 60× above NASH incidence. The model trained to detect NASH in the non-NAFL cohort had an AUPRC of 0.0030 (95% CI 0.0029 to 0.0031) and an AUROC of 0.78. At 10% recall, model precision was 1%, which is 20× above NASH incidence.

Conclusion

The low incidence of NASH in medical claims data corroborates the pattern of NASH underdiagnosis in clinical practice. Claims-based machine learning could facilitate the detection of patients with probable NASH for diagnostic testing and disease management.

Detecting non-alcoholic fatty liver disease and risk factors in health databases: accuracy and limitations of the ICD-10-AM

OBJECTIVE

The prevalence of non-alcoholic fatty liver disease (NAFLD) and non-alcoholic steatohepatitis (NASH) cirrhosis is often underestimated in healthcare and administrative databases that define disease burden using International Classification of Diseases (ICD) codes. This retrospective audit was conducted to explore the accuracy and limitations of the ICD, Tenth Revision, Australian Modification (ICD-10-AM) to detect NAFLD, metabolic risk factors (obesity and diabetes) and other aetiologies of chronic liver disease.

DESIGN/METHOD

ICD-10-AM codes in 308 admitted patient encounters at two major Australian tertiary hospitals were compared with data abstracted from patients' electronic medical records. Accuracy of individual codes and grouped combinations was determined by calculating sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and Cohen's kappa coefficient (κ).

RESULTS

The presence of an ICD-10-AM code accurately predicted the presence of NAFLD/NASH (PPV 91.2%) and obesity (PPV 91.6%) in most instances. However, codes underestimated the prevalence of NAFLD/NASH and obesity by 42.9% and 45.3%, respectively. Overall concordance between clinical documentation and 'grouped alcohol' codes (κ 0.75) and hepatitis C codes (κ 0.88) was high. Hepatitis B codes detected false-positive cases in patients with previous exposure (PPV 55.6%). Accuracy of codes to detect diabetes was excellent (sensitivity 95.8%; specificity 97.6%; PPV 94.9%; NPV 98.1%) with almost perfect concordance between codes and documentation in medical records (κ 0.93).

CONCLUSION

Recognition of the utility and limitations of ICD-10-AM codes to study the burden of NAFLD/NASH cirrhosis is imperative to inform public health strategies and appropriate investment of resources to manage this burgeoning chronic disease.