Diagnostic accuracy of ICD-9 code 780.2 for the identification of patients with syncope in the emergency department

Caveat emptor: The accuracy of claims data in appendicitis research

BACKGROUND

International Classification of Disease, ninth/tenth revision codes are used to identify patients with appendicitis and classify severity of disease for research and hospital reimbursement. We sought to determine accuracy of International Classification of Disease, ninth/tenth revision codes in classifying appendicitis as uncomplicated versus complicated (defined as perforated, necrotic, or abscess) compared with the clinical gold standard: surgeon characterization of the appendix in the operative report.

METHODS

Retrospective review of operative reports and discharge International Classification of Disease, ninth/tenth revision codes for patients ≥18 years old who underwent noninterval, nonincidental appendectomy between January 2012 and December 2019 at a tertiary referral center. Sensitivity, specificity, and positive predictive value were calculated for International Classification of Disease, ninth/tenth revision codes to classify appendicitis accurately as complicated compared with surgeon description. ICD-9/10 codes and surgeon description were categorized into complicated/uncomplicated based on the American Association for the Surgery of Trauma grading system.

RESULTS

In the study, 1,495 patients with acute appendicitis underwent appendectomy. Per surgeon description, 200 (13%) were complicated and 1,295 (87%) uncomplicated. Compared with surgeon description, discharge International Classification of Disease, ninth/tenth revision codes did not accurately identify complicated appendicitis: sensitivity = 0.68, positive predictive value = 0.77. As a sensitivity analysis, the cohort was stratified by public versus private payers, and the results did not change.

CONCLUSION

International Classification of Disease, ninth/tenth revision codes do not accurately identify surgeon-described complicated appendicitis. Nearly one-third of the cases of complicated appendicitis were coded as uncomplicated. Such misclassification negatively impacts reimbursement for complicated appendicitis care and could lead to misleading results in research and quality improvement activities that rely on these codes.

Comparative Accuracy of ICD-9 vs ICD-10 Codes for Acute Appendicitis

BACKGROUND

ICD codes are used to identify patients with appendicitis and to classify disease severity for reimbursement and research purposes. We sought to compare the accuracy of ICD-9 vs ICD-10 codes in classifying appendicitis as uncomplicated vs complicated (defined as perforated, necrotic, or abscess) compared with the clinical gold standard: surgeon characterization of the appendix in the operative report.

STUDY DESIGN

This is a retrospective review of operative reports and discharge ICD-9/10 codes for patients 18 years or older who underwent noninterval, nonincidental appendectomy from January 2012 to December 2019 at a tertiary referral center. Sensitivity, specificity, and positive predictive value were calculated for ICD-9/10 codes to classify appendicitis as complicated when compared with surgeon description. Chi-square testing was used to compare agreement between ICD-9/10 codes and surgeon description.

RESULTS

A total of 1,585 patients underwent appendectomy. ICD-9 codes had higher sensitivity than ICD-10 codes for complicated appendicitis (sensitivity 0.84 and 0.54, respectively) and a similar positive predictive value (0.77 and 0.76, respectively). Overall, 91% of ICD-9 codes agreed with surgical description of disease, but 84.4% of ICD-10 codes agreed with surgical description (p < 0.01). Among cases classified as complicated by the surgeon, 84% (79/94) had an accurate ICD-9 code for complicated disease, but only 53.8% (57/106) of cases had an accurate ICD-10 code (p < 0.01).

CONCLUSIONS

Compared with ICD-9 codes, ICD-10 codes were less accurate in characterizing severity of appendicitis. The ICD-10 coding schema does not provide an accurate representation of disease severity. Until this system is improved, significant caution is needed for people who rely on these data for billing, quality improvement, and research purposes.

United States ED Visits by Adult Women for Nonfatal Intimate Partner Strangulation, 2006 to 2014: Prevalence and Associated Characteristics

INTRODUCTION

Nonfatal intimate partner strangulation poses significant acute and long-term morbidity risks and also heightens women's risk for future femicide. The lifetime prevalence of nonfatal intimate partner strangulation has been estimated to be approximately 10%, or 11 million women, in the general United States population. Given the potential for significant health risks and serious consequences of strangulation, this study adds to the limited literature by estimating prevalence and describing the associated characteristics of strangulation-related visits among United States ED visits by adult women after intimate partner violence.

METHODS

Prevalence estimation as well as simple and multivariable logistic regression analyses were completed using data from the Nationwide Emergency Department Sample spanning the years 2006 to 2014.

RESULTS

The prevalence of strangulation codes was estimated at 1.2% of all intimate partner violence visits. Adjusting for visits, hospital characteristics, and visit year, higher odds of strangulation were noted in younger women, metropolitan hospitals, level I/II trauma centers, and non-Northeast regions. Increases in strangulation events among intimate partner violence-related visits in recent years were also observed.

DISCUSSION

A relatively low prevalence may reflect an underestimate of true nonfatal intimate partner strangulation visits owing to coding or a very low rate of ED visits for this issue. Higher odds of strangulation among intimate partner violence visits by women in more recent years may be due to increased recognition and documentation by frontline clinicians and coding teams. Continued research is needed to further inform clinical, postcare, and social policy efforts.

Association Between US Physician Malpractice Claims Rates and Hospital Admission Rates Among Patients With Lower-Risk Syncope

IMPORTANCE

The US Government Accountability Office has changed its estimate of the annual costs of defensive medicine, largely because it has been difficult to objectively measure its impact. Evaluating the association of malpractice claims rates with hospital admission rates and the costs of admitting patients with low-risk conditions would help to document the impact of defensive medicine. Although syncope is a concerning symptom, most patients with syncope have a low risk of adverse outcomes. However, many low-risk patients are still admitted to the hospital, with associated costs of more than $2.5 billion per year in the US.

OBJECTIVE

To assess whether hospital admission rates after emergency department visits among patients with lower-risk syncope are associated with state-level variations in malpractice claims rates.

DESIGN, SETTING, AND PARTICIPANTS

This cross-sectional study of emergency department visits among patients with lower-risk syncope used deidentified data from the Clinformatics Data Mart database (Optum). Lower-risk syncope visits were defined as those with a primary diagnosis of syncope and collapse based on International Classification of Diseases, Ninth Revision, Clinical Modification code 780.2 or International Classification of Diseases, Tenth Revision, Clinical Modification code R55 that did not include another major diagnostic code for a condition requiring hospital admission (such as heart disease, cancer, or medical shock) or an inpatient hospital stay of more than 3 days. These data were linked to publicly available data from the National Practitioner Data Bank pertaining to physician malpractice claims between January 1, 2008, and December 31, 2017. The 2 data sets were linked at the state-year level. Data were analyzed from October 2, 2019, to September 12, 2020.

MAIN OUTCOMES AND MEASURES

The association between the rate of hospital admission after emergency department visits among patients with lower-risk syncope and the rate of physician malpractice claims was assessed at the state-year level using a state-level fixed-effects model. Standardized costs obtained from the Clinformatics Data Mart database were adjusted for inflation and expressed in 2017 US dollars using the Consumer Price Index.

RESULTS

Among 40 482 813 emergency department visits between 2008 and 2017, 519 724 visits (1.3%) were associated with syncope. Of those, 234 750 visits (45.2%) met the criteria for lower-risk syncope. The mean (SD) age of patients in the lower-risk cohort was 71.8 (13.5) years; 141 050 patients (60.1%) were female, and 44 115 patients (18.8%) were admitted to the hospital, representing an extra cost of $6542 per admission. The mean rate of physician malpractice claims varied from 0.27 claims per 100 000 people to 8.63 claims per 100 000 people across states and across years within states. A state-level fixed-effects regression model indicated that, for every 1 in 100 000-person increase in the physician malpractice claims rate, there was an absolute increase of 6.70% (95% CI, 4.65%-8.75%) or a relative increase of 35.6% in the hospital admission rate, which represented an additional $102 million in costs associated with this lower-risk cohort.

CONCLUSIONS AND RELEVANCE

In this study, increases in physician malpractice claims rates were associated with increases in hospital admission rates and substantial health care costs for patients with lower-risk syncope, and these increases are likely associated with the practice of defensive medicine.

Incidence of Syncope During Pregnancy: Temporal Trends and Outcomes

Background

We examined temporal trends, timing, and frequency, as well as adverse neonatal and maternal outcomes occurring in the first year postpartum among women experiencing syncope during pregnancy.

Methods and Results

This was a retrospective study of pregnancies between January 1, 2005, and December 31, 2014, in the province of Alberta, Canada. Of 481 930 pregnancies, 4667 had an episode of syncope. Poisson regression analysis found a 5% increase/year (rate ratio, 1.05; 95% CI, 1.04–1.06) in the age‐adjusted incidence of syncope. Overall, 1506 (32.3%) of the syncope episodes first occurred in the first trimester, 2058 (44.1%) in the second trimester, and 1103 (23.6%) in the third trimester; and 8% (n=377) of pregnancies had >1 episode of syncope. Compared with women without syncope, women who experienced syncope were younger (age <25 years; 34.7% versus 20.8%; P<0.001), and primiparous (52.1% versus 42.4%; P<0.001). The rate of preterm birth was higher in pregnancies with syncope during the first trimester (18.3%), compared with the second (15.8%) and third trimesters (14.2%) and pregnancies without syncope (15.0%; P<0.01). The incidence of congenital anomalies among children born of pregnancies with multiple syncope episodes was significantly higher (4.9%) compared with children of pregnancies without syncope (2.9%; P<0.01). Within 1 year after delivery, women with syncope during pregnancy had higher rates of cardiac arrhythmias and syncope episodes than women with no syncope during pregnancy.

Conclusions

Pregnant women with syncope, especially when the syncopal event occurs during the first trimester, may be at a higher risk of adverse pregnancy outcomes as well as an increased incidence of cardiac arrhythmia and syncope postpartum.

Personalized risk stratification through attribute matching for clinical decision making in clinical conditions with aspecific symptoms: The example of syncope

BACKGROUND

Risk stratification is challenging in conditions, such as chest pain, shortness of breath and syncope, which can be the manifestation of many possible underlying diseases. In these cases, decision tools are unlikely to accurately identify all the different adverse events related to the possible etiologies. Attribute matching is a prediction method that matches an individual patient to a group of previously observed patients with identical characteristics and known outcome. We used syncope as a paradigm of clinical conditions presenting with aspecific symptoms to test the attribute matching method for the prediction of the personalized risk of adverse events.

METHODS

We selected the 8 predictor variables common to the individual-patient dataset of 5 prospective emergency department studies enrolling 3388 syncope patients. We calculated all possible combinations and the number of patients in each combination. We compared the predictive accuracy of attribute matching and logistic regression. We then classified ten random patients according to clinical judgment and attribute matching.

RESULTS

Attribute matching provided 253 of the 384 possible combinations in the dataset. Twelve (4.7%), 35 (13.8%), 50 (19.8%) and 160 (63.2%) combinations had a match size ≥50, ≥30, ≥20 and <10 patients, respectively. The AUC for the attribute matching and the multivariate model were 0.59 and 0.74, respectively.

CONCLUSIONS

Attribute matching is a promising tool for personalized and flexible risk prediction. Large databases will need to be used in future studies to test and apply the method in different conditions.

Artificial Intelligence Algorithms and Natural Language Processing for the Recognition of Syncope Patients on Emergency Department Medical Records

Background: Enrollment of large cohorts of syncope patients from administrative data is crucial for proper risk stratification but is limited by the enormous amount of time required for manual revision of medical records. Aim: To develop a Natural Language Processing (NLP) algorithm to automatically identify syncope from Emergency Department (ED) electronic medical records (EMRs). Methods: De-identified EMRs of all consecutive patients evaluated at Humanitas Research Hospital ED from 1 December 2013 to 31 March 2014 and from 1 December 2015 to 31 March 2016 were manually annotated to identify syncope. Records were combined in a single dataset and classified. The performance of combined multiple NLP feature selectors and classifiers was tested. Primary Outcomes: NLP algorithms’ accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and F3 score. Results: 15,098 and 15,222 records from 2013 and 2015 datasets were analyzed. Syncope was present in 571 records. Normalized Gini Index feature selector combined with Support Vector Machines classifier obtained the best F3 value (84.0%), with 92.2% sensitivity and 47.4% positive predictive value. A 96% analysis time reduction was computed, compared with EMRs manual review. Conclusions: This artificial intelligence algorithm enabled the automatic identification of a large population of syncope patients using EMRs.

Prevalence and Predictors of Pulmonary Embolism in Hospitalized Patients with Syncope

OBJECTIVES

Approximately one in six patients hospitalized with syncope have pulmonary embolism (PE), according to the PE in Syncope Italian Trial study. Subsequent studies using administrative data have reported a PE prevalence of <3%. The aim of the study was to determine the prevalence and predictors of PE in hospitalized patients with syncope.

METHODS

We retrospectively reviewed the records of patients who were hospitalized in the MedStar Washington Hospital Center between May 1, 2015 and June 30, 2017 with deep venous thrombosis, PE, and syncope. Only patients who presented to the emergency department with syncope were included in the final analysis. PE was diagnosed by either positive computed tomographic angiography or a high-probability ventilation-perfusion scan. Univariate and multivariate logistic regressions were used to assess the associations between clinical variables and the diagnosis of PE in patients with syncope.

RESULTS

Of the 408 patients hospitalized with syncope (mean age, 67.5 years; 51% men [N = 208]), 25 (6%) had a diagnosis of PE. Elevated troponin levels (odds ratio 6.6, 95% confidence interval 1.9-22.9) and a dilated right ventricle on echocardiogram (odds ratio 6.9, 95% confidence interval 2.0-23.6) were independently associated with the diagnosis of PE. Age, active cancer, and history of deep venous thrombosis were not associated with the diagnosis of PE.

CONCLUSIONS

The prevalence of PE in this study is approximately one-third of the reported prevalence in the PE in Syncope Italian Trial study and almost three times the value reported in administrative data-based studies. PE should be suspected in patients with syncope and elevated troponin levels or a dilated right ventricle on echocardiogram.