Explore the vast range of titles available.

Abstract Objectives Second-opinion pathology review identifies clinically significant diagnostic discrepancies for some patients. Discrepancy rates and laboratory-specific costs in a single health care system for patients referred from regional affiliates to a comprehensive cancer center (“main campus”) have not been reported. Methods Main campus second-opinion pathology cases for 740 patients from eight affiliated hospitals during 2016 to 2018 were reviewed. Chart review was performed to identify changes in care due to pathology review. To assess costs of pathology interpretation, reimbursement rates for consultation Current Procedural Terminology billing codes were compared with codes that would have been used had the cases originated at the main campus. Results Diagnostic discrepancies were identified in 104 (14.1%) patients, 30 (4.1%) of which resulted in a change in care. In aggregate, reimbursement for affiliate cases was 65.6% of the reimbursement for the same cases had they originated at the main campus. High-volume organ systems with low relative consultation reimbursement included gynecologic, breast, and thoracic. Conclusions Preventable diagnostic errors are reduced by pathology review for patients referred within a single health care system. Although the resulting changes in care potentially lead to overall cost savings, the financial value of referral pathology review could be improved.

Approximately 476,000 cases of Lyme disease are diagnosed in the United States annually, yet comprehensive economic evaluations are lacking. In a prospective study among reported cases in Lyme disease-endemic states, we estimated the total patient cost and total societal cost of the disease. In addition, we evaluated disease and demographic factors associated with total societal cost. Participants had a mean patient cost of ≈$1,200 (median $240) and a mean societal cost of ≈$2,000 (median $700). Patients with confirmed disseminated disease or probable disease had approximately double the societal cost of those with confirmed localized disease. The annual, aggregate cost of diagnosed Lyme disease could be $345-968 million (2016 US dollars) to US society. Our findings emphasize the importance of effective prevention and early diagnosis to reduce illness and associated costs. These results can be used in cost-effectiveness analyses of current and future prevention methods, such as a vaccine.

To determine the accuracy of recently introduced International Classification of Diseases Ninth Revision (ICD-9) procedure and Current Procedural Terminology (CPT) codes designed for injection or infusion of thrombolytic agents. We determined the accuracy of ICD-9 procedure code 99.10 and CPT codes 37201, 37202 for use of thrombolysis in ischemic stroke by comparing procedure codes of University Hospital discharge data with a concurrent prospective registry. Of the 369 ischemic stroke patients, 49 (13.3%) received either intravenous and/or intraarterial thrombolysis. The sensitivity and specificity for ICD-9 procedure code 99.10 was 55% and 98% and CPT procedure code 37201 and 37202 was 49% and 99%. Identification by either ICD-9 codes or CPT codes yielded a high sensitivity and specificity of 82% and 98%. The use of ICD-9 and CPT codes alone may underestimate the use of thrombolytics using national and regional database. Best results are achieved when a combination of ICD-9 and CPT codes are used to identify the use of thrombolytics.

Proper codification of medical diagnoses and procedures is essential for optimized health care management, quality improvement, research, and reimbursement tasks within large healthcare systems. Assignment of diagnostic or procedure codes is a tedious manual process, often prone to human error. Natural Language Processing (NLP) has been suggested to facilitate this manual codification process. Yet, little is known on best practices to utilize NLP for such applications. With Large Language Models (LLMs) becoming more ubiquitous in daily life, it is critical to remember, not every task requires that level of resource and effort. Here we comprehensively assessed the performance of common NLP techniques to predict current procedural terminology (CPT) from operative notes. CPT codes are commonly used to track surgical procedures and interventions and are the primary means for reimbursement. Our analysis of 100 most common musculoskeletal CPT codes suggest that traditional approaches can outperform more resource intensive approaches like BERT significantly (P-value = 4.4e-17) with average AUROC of 0.96 and accuracy of 0.97, in addition to providing interpretability which can be very helpful and even crucial in the clinical domain. We also proposed a complexity measure to quantify the complexity of a classification task and how this measure could influence the effect of dataset size on model’s performance. Finally, we provide preliminary evidence that NLP can help minimize the codification error, including mislabeling due to human error.

Study Design Retrospective Cohort Study. Objectives Using natural language processing (NLP) in combination with machine learning on standard operative notes may allow for efficient billing, maximization of collections, and minimization of coder error. This study was conducted as a pilot study to determine if a machine learning algorithm can accurately identify billing Current Procedural Terminology (CPT) codes on patient operative notes. Methods This was a retrospective analysis of operative notes from patients who underwent elective spine surgery by a single senior surgeon from 9/2015 to 1/2020. Algorithm performance was measured by performing receiver operating characteristic (ROC) analysis, calculating the area under the ROC curve (AUC) and the area under the precision-recall curve (AUPRC). A deep learning NLP algorithm and a Random Forest algorithm were both trained and tested on operative notes to predict CPT codes. CPT codes generated by the billing department were compared to those generated by our model. Results The random forest machine learning model had an AUC of .94 and an AUPRC of .85. The deep learning model had a final AUC of .72 and an AUPRC of .44. The random forest model had a weighted average, class-by-class accuracy of 87%. The LSTM deep learning model had a weighted average, class-by-class accuracy 0f 59%. Conclusions Combining natural language processing with machine learning is a valid approach for automatic generation of CPT billing codes. The random forest machine learning model outperformed the LSTM deep learning model in this case. These models can be used by orthopedic or neurosurgery departments to allow for efficient billing.

In 2023, changes were made to the Current Procedural Terminology (CPT) codes for anterior abdominal hernia repair to more uniformly reimburse hernia repair and better reflect current practices. These changes were made to address a shift toward the outpatient setting however general surgeons may be negatively impacted. A retroactive analysis of an ambulatory surgery center compared the surgeon's average reimbursement from old CPT codes from 2019 to 2022 to new CPT codes in 2023 including the evaluation and management (E/M) services in the new 0-day global period. Average case reimbursement to the surgeon decreased significantly for incarcerated hernia repair (p ​= ​0.01, −58.89 ​% change) and to the surgical facility for reducible hernia repair (p ​= ​0.004, −56.97 ​% change) between the combined average of 2019–2022 and 2023. Average procedural work relative value units for hernias from 2019 to 2022 were found to decrease by 25.4 ​% for incarcerated and 45 ​% for reducible hernias compared to 2023. Further evaluation with a larger surgical facility is needed to confirm these findings. •Average case reimbursement of incarcerated abdominal hernias decreased by 59 ​% in 2023 compared to 2019 to 2022 combined.•Change in reducible abdominal hernias repairs decreased by 44% but was not found to be statistically significant.•E/M reimbursement per case was about $165 in 2023.•In 2023, surgeon reimbursement for the repair alone is $248.97 for reducible and $272.70 for incarcerated abdominal hernias.•These CPT code changes are detrimental to surgeons performing abdominal wall hernia repairs.

Complex surgical specimens are associated with complex Current Procedural Terminology (CPT) coding. To assess and optimize the accuracy of CPT coding of complex genitourinary specimens at our institution. Baseline CPT codes for nephrectomy and cystectomy surgical pathology specimens were examined during a 3-month period. Pathology reports were reviewed for accurate CPT coding, and commensurate tests of change were implemented. Post-test-of-change data were re-collected, analyzed, and compared to the baseline data. Baseline data consisted of 71 genitourinary specimens (April to June 2021) and demonstrated undercoding in 46% (n = 33 of 71) of specimens, mostly in specimens with 2 or more billable organs. From findings in baseline data, we implemented test-of-change efforts consisting of awareness, education, and increased documentation and communication between all involved parties. Marked improvement was noted in the coding accuracy of specimens with 2 billable organs (pretest: n = 4 of 21, 19%; posttest: n = 14 of 21, 67%) and 3 or more billable organs (pretest: n = 0 of 16, 0%; posttest: n = 7 of 12, 58%) (P value = .002). Problematic areas included nephrectomy specimens resected with adrenal glands (pretest: n = 2 of 12, 17%; posttest: n = 12 of 14, 86%) and ureters for urothelial carcinoma (pretest: n = 0 of 10, 0%; posttest: n = 3 of 6, 50%), as well as regional lymph nodes commingled with resection specimens (pretest: n = 0 of 11, 0%; posttest: n = 7 of 9, 78%). A comprehensive approach involving all stakeholders is necessary for CPT coding of complex surgical specimens. Documentation and familiarity with coding rules, specifically bundling and unbundling, as well as clinical indications for resection, are important factors in optimizing CPT coding.

The American College of Surgeons National Surgical Quality Improvement Project (ACS-NSQIP) uses Current Procedural Terminology (CPT) codes for risk-adjusted calculations. This study evaluates the inter-rater reliability of coding colorectal resections across Canada by ACS-NSQIP surgical clinical nurse reviewers (SCNR) and its impact on risk predictions. SCNRs in Canada were asked to code simulated operative reports. Percent agreement and free-marginal kappa correlation were calculated. The ACS-NSQIP risk calculator was utilized to illustrate its impact on risk prediction. Responses from 44 of 150 (29.3 ​%) SCNRs revealed 3 to 6 different codes chosen per case, with agreement ranging from 6.7 ​% to 62.3 ​%. Free-marginal kappa correlation ranged from moderate agreement (0.53) to high disagreement (−0.17). ACS-NSQIP risk calculator predicted large absolute differences in risk for serious complications (0.2 ​%–13.7 ​%) and mortality (0.2 ​%–6.3 ​%). This study demonstrated low inter-rater reliability in coding ACS-NSQIP colorectal procedures in Canada among SCNRs, impacting risk predictions. •This study demonstrated low inter-rater reliability across all ACS-NSQIP colorectal resection procedure coding in Canada among SCNRs.•Coding inconsistencies resulted in significant variation in predicted morbidity and mortality, when using the NSQIP risk prediction calculator.•These results highlight the importance of ongoing efforts to improve coding standardization and education among healthcare professionals.

INTRODUCTION:Rising healthcare costs have led to decreasing reimbursements for various procedures and providers. We chose to analyze Medicare reimbursement trends for 26 esophagogastroduodenoscopy (EGD) and 31 colonoscopy current procedural terminology (CPT) codes from 2018 to 2023 for hospital outpatient centers, ambulatory surgical centers (ASC), and gastroenterologists. We also wanted to look at the effects of inflation on these Medicare reimbursements.METHODS:We calculated the nominal percentage change from 2018 to 2023 for each CPT code. We then took inflation data provided by the US Bureau of Labor Statistics and calculated the real change in reimbursements from 2018 to 2023 for each of the 31 colonoscopy and 26 EGD CPT codes.RESULTS:Our results show that although nominal reimbursements to physicians have been steadily declining for performing gastrointestinal procedures, nominal reimbursements to hospital outpatient and ASC have been increasing from 2018 to 2023. After taking into account inflation, physicians saw significant decreases in real purchasing power for performing EGD and colonoscopies. ASC and hospital outpatient centers saw reimbursements keep up with inflation.DISCUSSION:Physician reimbursements for gastroenterology procedures make up a small portion of reimbursements by Medicare compared to Medicare reimbursements to facilities such as ASC and hospital outpatient centers. However, physicians have seen significant reimbursement cuts, whereas facilities have not. Moreover, higher inflation leads to increased expenses for gastroenterology practices. It remains to be seen how these reimbursement changes will affect patients access to care and physicians practice sustainability.

To assess strategies for enhancing the generalizability of healthcare artificial intelligence models, we analyzed the impact of preprocessing approaches applied to medical free text, compared single- versus multiple-institution data models, and evaluated data divergence metrics. From 1,607,393 procedures across 44 U.S. institutions, deep neural network models were created to classify anesthesiology Current Procedural Terminology codes from medical free text. Three levels of text preprocessing were analyzed from minimal to automated (cSpell) with comprehensive physician review. Kullback–Leibler Divergence and k-medoid clustering were used to predict single- vs multiple-institutional model performances. Single-institution models showed a mean accuracy of 92.5% [2.8% SD] and 0.923 [0.029] F1 on internal data but generalized poorly on external data (− 22.4% [7.0%]; − 0.223 [0.081]). Free text preprocessing minimally altered performance (+ 0.51% [2.23]; + 0.004 [0.020]). An all-institution model performed worse on internal data (-4.88% [2.43%]; − 0.045 [0.020]), but improved generalizability to external data (+ 17.1% [8.7%]; + 0.182 [0.073]). Compared to vocabulary overlap and Jaccard similarity, Kullback–Leibler Divergence correlated with model performance (R 2 of 0.41 vs 0.16 vs 0.08, respectively) and was successful clustering institutions and identifying outlier data. Overall, pre-processing medical free text showed limited utility improving generalization of machine learning models, single institution models performed best but generalized poorly, while combined data models improved generalization but never achieved performance of single-institutional models. Kullback–Leibler Divergence provided valuable insight as a reliable heuristic to evaluate generalizability. These results have important implications in developing broad use artificial intelligence healthcare applications, providing valuable insight into their development and evaluations.