Clinical text classification with rule-based features and knowledge-guided convolutional neural networks.

BACKGROUND: Clinical text classification is an fundamental problem in medical natural language processing. Existing studies have cocnventionally focused on rules or knowledge sources-based feature engineering, but only a limited number of studies have exploited effective representation learning capability of deep learning methods. METHODS: In this study, we propose a new approach which combines rule-based features and knowledge-guided deep learning models for effective disease classification. Critical Steps of our method include recognizing trigger phrases, predicting classes with very few examples using trigger phrases and training a convolutional neural network (CNN) with word embeddings and Unified Medical Language System (UMLS) entity embeddings. RESULTS: We evaluated our method on the 2008 Integrating Informatics with Biology and the Bedside (i2b2) obesity challenge. The results demonstrate that our method outperforms the state-of-the-art methods. CONCLUSION: We showed that CNN model is powerful for learning effective hidden features, and CUIs embeddings are helpful for building clinical text representations. This shows integrating domain knowledge into CNN models is promising.

Early impact of the ICD-10-CM transition on selected health outcomes in 13 electronic health care databases in the United States.

PURPOSE: To describe the consistency in the frequency of 5 health outcomes across the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) and Tenth Revision, Clinical Modification (ICD-10-CM) eras in the US. METHODS: We examined the incidence of 3 acute conditions (acute myocardial infarction [AMI], angioedema, ischemic stroke) and the prevalence of 2 chronic conditions (diabetes, hypertension) during the final 5 years of the ICD-9-CM era (January 2010-September 2015) and the first 15 months of the ICD-10-CM era (October 2015-December 2016) in 13 electronic health care databases in the Sentinel System. For each health outcome reviewed during the ICD-10-CM era, we evaluated 4 definitions, including published algorithms derived from other countries, as well as simple-forward, simple-backward, and forward-backward mapping using the General Equivalence Mappings. For acute conditions, we also compared the incidence between April to December 2014 and April to December 2016. RESULTS: The analyses included data from approximately 172 million health plan members. While the incidence or prevalence of AMI and hypertension performed similarly across the 2 eras, the other 3 outcomes did not demonstrate consistent trends for some or all the ICD-10-CM definitions assessed. CONCLUSIONS: When using data from both the ICD-9-CM and ICD-10-CM eras, or when using results from ICD-10-CM data to compare to results from ICD-9-CM data, researchers should test multiple ICD-10-CM outcome definitions as part of sensitivity analysis. Ongoing assessment of the impact of ICD-10-CM transition on identification of health outcomes in US electronic health care databases should occur as more data accrue.

Assessing the impact of the new ICD-10-CM coding system on pharmacoepidemiologic studies-An application to the known association between angiotensin-converting enzyme inhibitors and angioedema.

PURPOSE: To replicate the well-established association between angiotensin-converting enzyme inhibitors versus beta blockers and angioedema in the International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) era. METHODS: We conducted a retrospective, inception cohort study in a large insurance database formatted to the Sentinel Common Data Model. We defined study periods spanning the ICD-9-CM era only, ICD-10-CM era only, and ICD-9-CM and ICD-10-CM era and conducted simple-forward mapping (SFM), simple-backward mapping (SBM), and forward-backward mapping (FBM) referencing the General Equivalence Mappings to translate the outcome (angioedema) and covariates from ICD-9-CM to ICD-10-CM. We performed propensity score (PS)-matched and PS-stratified Cox proportional hazards regression to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). RESULTS: In the ICD-9-CM and ICD-10-CM eras spanning April 1 to September 30 of 2015 and 2016, there were 152 017 and 145 232 angiotensin-converting enzyme inhibitor initiators and 115 073 and 116 652 beta-blocker initiators, respectively. The PS-matched HR was 4.19 (95% CI, 2.82-6.23) in the ICD-9-CM era, 4.37 (2.92-6.52) in the ICD-10-CM era using SFM, and 4.64 (3.05-7.07) in the ICD-10-CM era using SBM and FBM. The PS-matched HRs from the mixed ICD-9-CM and ICD-10-CM eras ranged from 3.91 (2.69-5.68) to 4.35 (3.33-5.70). CONCLUSION: The adjusted HRs across different diagnostic coding eras and the use of SFM versus SBM and FBM produced numerically different but clinically similar results. Additional investigations as ICD-10-CM data accumulate are warranted.

Uso inadecuado de términos vasculares: mitos y leyendas / Inadequate use of vascular ends: Challenges and legends

No disponible

Elective ambulatory surgical care in Ireland-why it needs to be better coded, classified and managed.

BACKGROUND: In the year to July 2017, surgical disciplines accounted for 73% of the total national inpatient and day case waiting list and, of these, day cases accounted for 72%. Their proper classification is therefore important so that patients can be managed and treated in the most suitable and efficient setting. AIMS: We set out to sub-classify the different elective surgical day cases treated in Irish public hospitals in order to assess their need to be managed as day cases and the consistency of practice between hospitals. METHODS: We analysed all elective day cases that came under the care of surgeons between January 2014 and December 2016 and sub-classified them into those that were (A) true day case surgical procedures; (B) minor surgery or outpatient procedures; (C) gastrointestinal endoscopies; (D) day case, non-surgical interventions and (E) unclassified or having no primary procedure identified. RESULTS: Of 813,236 day case surgical interventions performed over 3 years, 26% were adjudged to accord with group A, 41% with B, 23% with C, 5% with D and 5% with E. The ratio of A to B procedures did not vary significantly across the range of hospital types. However, there were some notable variations in coding and practices between hospitals. CONCLUSION: Our findings show that many day cases should have been performed as outpatient procedures and that there were variations in coding and practices between hospitals that could not be easily explained. Outpatient procedure coding and a better, more consistent, classification of day cases are both required to better manage this group of patients.

Variations in Medicare Reimbursement in Radiation Oncology: An Analysis of the Medicare Provider Utilization and Payment Data Set.

PURPOSE: The purposes of this study were to summarize recently published data on Medicare reimbursement to individual radiation oncologists and to identify the causes of variation in Medicare reimbursement in radiation oncology. METHODS AND MATERIALS: The Medicare Provider Utilization and Payment Data: Physician and Other Supplier Public Use File (POSPUF), which details nearly all services provided by radiation oncologists in 2012, was used for this study. The data were filtered and analyzed by physician and by billing code. Statistical analysis was performed to identify differences in reimbursements based on sex, rurality, billing of technical services, or location in a certificate of need (CON) state. RESULTS: There were 4135 radiation oncologists who received a total of $1,499,625,803 in payments from Medicare in 2012. Seventy-five percent of radiation oncologists were male. The median reimbursement was $146,453. The code with the highest total reimbursement was 77418 (radiation treatment delivery intensity modulated radiation therapy [IMRT]). The most commonly billed evaluation and management (E/M) code for new visits was 99205 (49%). The most commonly billed E/M code for established visits was 99213 (54%). Forty percent of providers billed none of their new office visits using 99205 (the highest E/M billing code), whereas 34% of providers billed all of their new office visits using 99205. For the 1510 radiation oncologists (37%) who billed technical services, median Medicare reimbursement was $606,008, compared with $93,921 for all other radiation oncologists (P<.001). On multivariate analysis, technical services billing (P<.001), male sex (P<.001), and rural location (P=.007) were predictive of higher Medicare reimbursement. CONCLUSIONS: The billing of technical services, with their high capital and labor overhead requirements, limits any comparison in reimbursement between individual radiation oncologists or between radiation oncologists and other specialists. Male sex and rural practice location are independent predictors of higher total Medicare reimbursements.

Correct coding for laboratory procedures during assisted reproductive technology cycles.

This document provides updated coding information for services related to assisted reproductive technology procedures. This document replaces the 2012 ASRM document of the same name.

Simulation of ICD-9 to ICD-10-CM Transition for Family Medicine: Simple or Convoluted?

OBJECTIVE: The objective of this study was to examine the impact of the transition from International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM), to Interactional Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM), on family medicine and to identify areas where additional training might be required. METHODS: Family medicine ICD-9-CM codes were obtained from an Illinois Medicaid data set (113,000 patient visits and $5.5 million in claims). Using the science of networks, we evaluated each ICD-9-CM code used by family medicine physicians to determine whether the transition was simple or convoluted. A simple transition is defined as 1 ICD-9-CM code mapping to 1 ICD-10-CM code, or 1 ICD-9-CM code mapping to multiple ICD-10-CM codes. A convoluted transition is where the transitions between coding systems is nonreciprocal and complex, with multiple codes for which definitions become intertwined. Three family medicine physicians evaluated the most frequently encountered complex mappings for clinical accuracy. RESULTS: Of the 1635 diagnosis codes used by family medicine physicians, 70% of the codes were categorized as simple, 27% of codes were convoluted, and 3% had no mapping. For the visits, 75%, 24%, and 1% corresponded with simple, convoluted, and no mapping, respectively. Payment for submitted claims was similarly aligned. Of the frequently encountered convoluted codes, 3 diagnosis codes were clinically incorrect, but they represent only <0.1% of the overall diagnosis codes. CONCLUSIONS: The transition to ICD-10-CM is simple for 70% or more of diagnosis codes, visits, and reimbursement for a family medicine physician. However, some frequently used codes for disease management are convoluted and incorrect, and for which additional resources need to be invested to ensure a successful transition to ICD-10-CM.

Coding of Neuroinfectious Diseases.

Accurate coding is an important function of neurologic practice. This contribution to Continuum is part of an ongoing series that presents helpful coding information along with examples related to the issue topic. Tips for diagnosis coding, Evaluation and Management coding, procedure coding, or a combination are presented, depending on which is most applicable to the subject area of the issue.

What explains DRG upcoding in neonatology? The roles of financial incentives and infant health.

We use the introduction of diagnosis related groups (DRGs) in German neonatology to study the determinants of upcoding. Since 2003, reimbursement is based inter alia on birth weight, with substantial discontinuities at eight thresholds. These discontinuities create incentives to upcode preterm infants into classes of lower birth weight. Using data from the German birth statistics 1996-2010 and German hospital data from 2006 to 2011, we show that (1) since the introduction of DRGs, hospitals have upcoded at least 12,000 preterm infants and gained additional reimbursement in excess of 100 million Euro; (2) upcoding rates are systematically higher at thresholds with larger reimbursement hikes and in hospitals that subsequently treat preterm infants, i.e. where the gains accrue; (3) upcoding is systematically linked with newborn health conditional on birth weight. Doctors and midwives respond to financial incentives by not upcoding newborns with low survival probabilities, and by upcoding infants with higher expected treatment costs.

Developing and modifying behavioral coding schemes in pediatric psychology: a practical guide.

OBJECTIVES: To provide a concise and practical guide to the development, modification, and use of behavioral coding schemes for observational data in pediatric psychology. METHODS: This article provides a review of relevant literature and experience in developing and refining behavioral coding schemes. RESULTS: A step-by-step guide to developing and/or modifying behavioral coding schemes is provided. Major steps include refining a research question, developing or refining the coding manual, piloting and refining the coding manual, and implementing the coding scheme. Major tasks within each step are discussed, and pediatric psychology examples are provided throughout. CONCLUSIONS: Behavioral coding can be a complex and time-intensive process, but the approach is invaluable in allowing researchers to address clinically relevant research questions in ways that would not otherwise be possible.

ICD-10: are you ready for a brave new world?

The ICD-10 transition will be an evolutionary process. Relying on the EHR or certified coding staff alone will not be sufficient. The EHR can facilitate easy search tools that assist the provider in selecting a diagnosis. Billing staff are an invaluable resource to help validate that coding and documentation are in sync but the burden will clearly rest on the provider. The provider will be juggling a new code structure, drilling down to new levels of complexity and ensuring their documentation supports the specificity of the new codes selected, all while managing a full patient schedule. Education for the provider will be of paramount importance as they navigate this brave new world.

Discovering health knowledge in the BC nurse practitioners encounter codes.

Integrating the Nurse Practitioner (NP) role into clinical practice settings is new in British Columbia (BC), Canada. Encounter codes are unique numeric codes assigned to specific types of patient care services performed by NPs. In this study we apply knowledge discovery techniques to analyze the encounter codes extracted from the BC Ministry of Health database to understand the most common practice activities carried out by NPs and what diseases patients sought care for from NPs. The analysis produced important information about NPs' practice patterns. This work leads to a better understanding of NP practice patterns in BC.

Ethical and educational considerations in coding hand surgeries.

PURPOSE: To assess treatment coding knowledge and practices among residents, fellows, and attending hand surgeons. METHODS: Through the use of 6 hypothetical cases, we developed a coding survey to assess coding knowledge and practices. We e-mailed this survey to residents, fellows, and attending hand surgeons. In additionally, we asked 2 professional coders to code these cases. RESULTS: A total of 71 participants completed the survey out of 134 people to whom the survey was sent (response rate = 53%). We observed marked disparity in codes chosen among surgeons and among professional coders. CONCLUSIONS: Results of this study indicate that coding knowledge, not just its ethical application, had a major role in coding procedures accurately. Surgical coding is an essential part of a hand surgeon's practice and is not well learned during residency or fellowship. Whereas ethical issues such as deliberate unbundling and upcoding may have a role in inaccurate coding, lack of knowledge among surgeons and coders has a major role as well. CLINICAL RELEVANCE: Coding has a critical role in every hand surgery practice. Inconstancies among those polled in this study reveal that an increase in education on coding during training and improvement in the clarity and consistency of the Current Procedural Terminology coding rules themselves are needed.

Orthopedics coding and funding.

The French tarification à l'activité (T2A) prospective payment system is a financial system in which a health-care institution's resources are based on performed activity. Activity is described via the PMSI medical information system (programme de médicalisation du système d'information). The PMSI classifies hospital cases by clinical and economic categories known as diagnosis-related groups (DRG), each with an associated price tag. Coding a hospital case involves giving as realistic a description as possible so as to categorize it in the right DRG and thus ensure appropriate payment. For this, it is essential to understand what determines the pricing of inpatient stay: namely, the code for the surgical procedure, the patient's principal diagnosis (reason for admission), codes for comorbidities (everything that adds to management burden), and the management of the length of inpatient stay. The PMSI is used to analyze the institution's activity and dynamism: change on previous year, relation to target, and comparison with competing institutions based on indicators such as the mean length of stay performance indicator (MLS PI). The T2A system improves overall care efficiency. Quality of care, however, is not presently taken account of in the payment made to the institution, as there are no indicators for this; work needs to be done on this topic.

Development of an algorithm for identifying rheumatoid arthritis in the Korean National Health Insurance claims database.

This study aimed to develop an identification algorithm for validating the International Classification of Diseases-Tenth diagnostic codes for rheumatoid arthritis (RA) in the Korean National Health Insurance (NHI) claims database. An individual copayment beneficiaries program for rare and intractable diseases, including seropositive RA (M05), began in South Korea in July 2009. Patients registered in this system pay only 10 % of their total medical costs, but registration requires an official report from a doctor documenting that the patient fulfills the 1987 ACR criteria. We regarded patients registered in this system as gold standard RA and examined the validity of several algorithms to define RA diagnosis using diagnostic codes and prescription data. We constructed nine algorithms using two highly specific prescriptions (positive predictive value >90 % and specificity >90 %) and one prescription with high sensitivity (>80 %) and accuracy (>75 %). A total of 59,823 RA patients were included in this validation study. Among them, 50,082 (83.7 %) were registered in the individual copayment beneficiaries program and considered true RA. We tested nine algorithms that incorporated two specific regimens [biologics and leflunomide alone, methotrexate plus leflunomide, or more than 3 disease-modifying anti-rheumatic drugs (DMARDs)] and one sensitive drug (any non-steroidal anti-inflammatory drug (NSAID), any DMARD, or any NSAID plus any DMARD). The algorithm that included biologics, more than 3 DMARDs, and any DMARD yielded the highest accuracy (91.4 %). Patients with RA diagnostic codes with prescription of biologics or any DMARD can be considered as accurate cases of RA in Korean NHI claims database.