Diagnosis code assignment is a field that looks at automatically assigning diagnosis codes to free-text clinical notes. Assigning a diagnosis code to clinical notes manually needs expertise and time. Being able to do this automatically makes getting structured data from free-text clinical notes in Electronic Health Records easier. Furthermore, it can also be used as decision support for clinicians where they can input their notes and get back diagnosis codes as a second opinion. This project investigates the effects of using the hierarchies the diagnosis codes are structured in when training the diagnosis code assignment models compared to models trained with a standard loss function, binary cross-entropy. This has been done by using the hierarchy of two systems of diagnosis codes, ICD-9 and SNOMED CT, where one hierarchy is more detailed than the other. The results showed that hierarchical training increased the recall of the models regardless of what hierarchy was used. The more detailed hierarchy, SNOMED CT, increased the recall more than what the use of the less detailed ICD-9 hierarchy did. However, when using the more detailed SNOMED CT hierarchy the precision of the models decreased while the differences in precision when using the ICD-9 hierarchy was not statistically significant. The increase in recall did not make up for the decrease in precision when training with the SNOMED CT hierarchy when looking at the F1-score that is the harmonic mean of the two metrics. The conclusions from these results are that using a more detailed hierarchy increased the recall of the model more than when using a less detailed hierarchy. However, the overall performance measured in F1-score decreased when using a more detailed hierarchy since the other metric, precision, decreased by more than what recall increased. The use of a less detailed hierarchy maintained its precision giving an increase in overall performance. / Diagnoskodstilldeling är ett fält som undersöker hur man automatiskt kan tilldela diagnoskoder till fri-text läkaranteckningar. En manuell tildeling kräver expertis och mycket tid. Förmågan att göra detta automatiskt förenklar utvinning av strukturerad data från fri-text läkaranteckningar i elektroniska patientjournaler. Det kan även användas som ett hjälpverktyg för läkare där de kan skriva in sina läkaranteckningar och få tillbaka diagnoskoder som en andra åsikt. Detta arbete undersöker effekterna av att ta användning av hierarkierna diagnoskoderna är strukturerade i när man tränar modeller för diagnoskodstilldelning jämfört med att träna modellerna med en vanlig loss-funktion. Det här kommer att göras genom att använda hierarkierna av två diagnoskod-system, SNOMED CT och ICD-9, där en av hierarkierna är mer detaljerad. Resultaten visade att hierarkisk träning ökade recall för modellerna med båda hierarkierna. Den mer detaljerade hierarkien, SNOMED CT, gav en högre ökning än vad träningen med ICD-9 gjorde. Trots detta minskade precision av modellen när man den tränades med SNOMED CT hierarkin medan skillnaderna i precision när man tränade hierarkiskt med ICD-9 jämfört med vanligt inte var statistiskt signifikanta. Ökningen i recall kompenserade inte för minskningen i precision när modellen tränades med SNOMED CT hierarkien som man kan see på F1-score vilket är det harmoniska medelvärdet av de recall och precision. Slutsatserna man kan dra från de här resultaten är att en mer detaljerad hierarki kommer att öka recall mer än en mindre detaljerad hierarki ökar recall. Trots detta kommer den totala prestandan, som mäts av F1-score, försämras med en mer detaljerad hierarki eftersom att recall minskar mer än vad precision ökar. En mindre detaljerad hierarki i träning kommer bibehålla precision så att dens totala prestandan förbättras.

The Emergency department at Södersjukhuset currently suffers from very long waiting times. This is partly due to problems within visualisation and mapping of patient data and other information that is fundamental in the handling of patients at the Emergency department. This led to a need in the creation of improvement suggestions to the visualisation of the patient flow between the Emergency department and the Gastroenterology department at Södersjukhuset. During the project, a simulated graphical user interface was created with the purpose of mimicking Södersjukhusets current patient flow. This simulated user interface would visualise the patient flow between the Emergency department and the Gastroenterology department. Additionally, a patient symptoms estimation algorithm was implemented to guess the likelihood of a patient being admitted to a department. The result shows that there are many possible improvements to Södersjukhusets current hospital information system, TakeCare, that would facilitate the care coordinators work and in turn lower the waiting times at the Emergency department. / Akutmottagningen på Södersjukhuset har i dagsläget väldigt långa väntetider. Detta beror till viss del utav problem inom visualiseringen och kartläggning av patient data och annan fundamental information för att hantera patienter på akutmottagningen. Detta ledde till att det finns ett behov att skapa förbättringsförslag på visualiseringen av patientflödet mellan akutmottagningen och gastroenterologiavdelningen på Södersjukhuset. Under projektets gång skapades ett simulerat användargränssnitt med syfte att efterlikna Södersjukhusets nuvarande patientflöde. Denna lösning visualiserar patientflödet mellan akutmottagningen och gastroenterologiavdelningen. Dessutom implementerades en enkel sorteringsalgoritm som kan bedöma sannolikheten om en patient skall bli inlagd på en avdelning. Resultatet visar att det finns flera möjliga förbättringar i Södersjukhusets nuvarande elektroniska journalsystemet, TakeCare, som skulle underlätta vårdkoordinatorernas arbete och därmed sänka väntetiderna på akutmottagningen.

Consensus is lacking on early diagnostic criteria and the exact symptoms of Alzheimer's disease (AD). A new, in-office test may help physicians detect the early symptoms of AD, based upon new National Institute of Aging (NIA) criteria. However, a gap exists in knowledge regarding physicians' current use or intent to use the new protocols. Choreographing the descriptive AD terminology in the Diagnostic and Statistical Manual of Mental Disorders IV-TR and the International Classification of Diseases (ICD-10) is recommended. Thus, the purpose of this study was to understand possible contributing factors to physician's use or intent to use of the new NIA's diagnostic protocol. Data collected from 55 clinicians within 2 Northern California counties were analyzed using a bivariate test. The 2 dependent variables were physicians' use of, or intent to use, the NIA protocol; the 6 independent variables were number of years since graduating from medical school, area of specialty, percentage of patients over age 60 years, physician's gender, age, and knowledge about AD, as indicated by performance on the Alzheimer's Disease Knowledge Scale. The results of regression analyses indicated no statistical significant associations between the variables of interest (p = or greater than .05). This study is a first attempt at understanding physician attitudes toward, and usage patterns of, an important new in-office tool for early detection of AD. Further research using a larger sample size to increase power is needed. These findings have implications for positive social change by promoting an earlier detection of Alzheimer's disease, underscoring the need for additional training, and revising the terminology used in clinicians' desktop references.

No description available.

Magister Public Health - MPH / Background: The Provincial Government Western Cape (PGWC) Department of Health identified a lack of data on inpatient diagnoses and procedures in a form suitable to use for operational, strategic as well as financial health care planning. The only format in which diagnostic and procedure data was available was a paper based one encompassing individual patient notes in folders and discharge summaries. Making the data available in a coded format within an electronic database would facilitate storage, analysis and utilisation of that data for health service planning. Recognising the lack of availability of such coded data, this study was undertaken to evaluate different coding systems for their ability to code data in order to assist in deciding which coding systems best fit the need to facilitate easy and accurate recording of data on diagnoses and procedures from patient records. The identification of the most appropriate coding system for the context in which the PGWC Department of health functions should facilitate the easy recording, storage and retrieval of data that is accurate, reliable and useful for management decision making and would support optimal patient care. Aim: The aim of the study was to evaluate a selection of potentially suitable coding systems in order to determine which would be best able to code public sector district and regional hospital diagnostic and procedure data in the Western Cape Province. Method: A cross sectional analytical study design was used. Discharge diagnosis and procedure data were extracted from 342 patient folders from 3 district and 3 regional public hospitals in the Western Cape. This yielded 221 different diagnostic concepts and 126 different procedure concepts. The diagnostic concepts were further grouped into “all” diagnostic concepts recorded, diagnostic concepts recorded as “symptoms only” and diagnostic concepts recorded as “proper diagnoses”. The diagnostic coding systems evaluated were ICD-10 (International Classification of Diseases), ICPC-2 (International Classification of Primary Care 2nd edition) and ICD-10 Condensed Morbidity List. The procedure coding systems evaluated were CCSA-2001 (Current Procedure Terminology for South Africa) ICD-9-CM (International Classification of Diseases Clinical Modification 9th revision) and ICPC-2. The diagnoses and procedures were then coded in all of the coding systems being evaluated. Each diagnosis and procedure concept was matched with its representing concept in the coding system and scored according to the ability of the coding system to provide an “exact” match which was scored as (3) or a “partial” match scored as (2) or a “poor” match scored as (1) or “no” match scored as (0). Results: ICD-10 was better able to code diagnoses obtained from district and regional hospitals in the Western Cape compared to ICPC-2 and ICD-10 Condensed Morbidity list. For all recorded diagnostic concepts, ICD-10 was able to score 82% of the concepts as either an “exact” or a “partial” match compared to 79% in ICPC-2 and 30% in ICD-10-CL. ICD-10 consistently performed best across different stratification of diagnostic concepts namely concepts recorded as “proper diagnoses”, concepts recorded from regional hospitals only, concepts recorded from district hospitals only, concepts designated as “common diagnoses” and for concepts designated as “very common diagnoses”. In addition ICD-10 had zero diagnostic concepts for which “no match” could be found. CCSA -2001 proved to be the best coding system for coding procedures across all hospitals with an overall percentage of “exact” and “partial” matches of 83% compared to 65% for ICD-9-CM and 39% for ICPC-2 and also proved to be best across all strata. Conclusion: There were striking differences between the evaluated coding systems with regard to their ability to code diagnoses and procedures in the evaluated district and regional hospitals in the Western Cape Province. ICD-10 covers the scope of clinical diagnoses in more accurate and specific detail than ICPC-2 and ICD-10 CL. Though ICPC-2 is simpler and easier to use than ICD-10, it is not as detailed and specific as the latter but it proved ideal for symptoms rather than for specific diagnoses. ICD-10 Condensed Morbidity List was shown to be inadequate for coding diagnoses. However the difference between the two, although statistically significant were not very large and given the ease of use of ICPC-2, it could be recommended for use. As for procedures CCSA-2001 was assessed as being the most appropriate for coding procedures recorded in this setting compared to the other coding systems. ICPC-2 performed poorest for coding procedures across all evaluated settings and thus would be inappropriate to use. ICD-10 in most comparisons performed second best to ICPC-2 in terms of coding ability for diagnoses and could be considered for recommendation as a diagnostic coding tool.

The work group editing trauma disorders for the upcoming edition of the International Classification of Diseases (ICD-11) made several changes. Specifically, they significantly simplified the guidelines for post-traumatic stress disorder (PTSD) and added a new trauma disorder called complex PTSD (CPTSD). The new domains for PTSD and the addition of CPTSD require new instruments to assess these novel constructs. We developed a measure of PTSD and CPTSD (Complex Trauma Inventory; CTI) according to the proposed ICD-11 domains, creating several items to assess each domain. We examined the factor structure of the CTI (using both exploratory and confirmatory factor analyses) in two separate samples of diverse college students (n1 = 501; n2 = 500), reducing the original 53 trauma items in the item pool to 21 items. Confirmatory factor analyses supported two highly-correlated second-order factors (PTSD and complex factors), with PTSD (i.e., re-experiencing, avoidance, hyper-arousal) and complex factors (i.e., affect dysregulation, alterations in self-perception and alterations in relationships with others) each loading on three of the six ICD-11-consistent first-order factors (RMSEA = .08, CFI = .92, GFI = .87, SRMR = .06). Internal consistency for PTSD (α = .92) and complex factors (α = .93) are excellent.

Millions of people in the United States alone suffer from undiagnosed or late-diagnosed chronic diseases such as Chronic Kidney Disease and Type II Diabetes. Catching these diseases earlier facilitates preventive healthcare interventions, which in turn can lead to tremendous cost savings and improved health outcomes. We develop algorithms for predicting disease occurrence by drawing from ideas and techniques in the field of machine learning. We explore standard classification methods such as logistic regression and random forest, as well as more sophisticated sequence models, including recurrent neural networks. We focus especially on the use of medical code data for disease prediction, and explore different ways for representing such data in our prediction algorithms.

The number of annual implantable cardioverter defibrillator (ICD) implants has substantially increased over the last 5 years and is expected to grow rapidly. Implantable cardioverter defibrillators have a proven mortality benefit by terminating the life-threatening arrhythmias, even near end of life. In patients with moderate/severe symptomatic heart failure, enough clinical literature representing mortality benefits has been published, but limited numbers of studies have reviewed the dwindling risk-benefit profile near end of life, studying quality of life (QoL)/psychosocial impact. Criteria outlining either continued use or deactivation policy/procedures near end of life have not been clearly defined and/or largely implemented, which in turn requires more focused research using multifactorial approach to determine improved patient-centered outcomes.

Clinical documentation is at the center of a patient's medical record; this record contains all the information applicable to the care a patient receives in the hospital. The practice problem addressed in this project was the lack of clear, consistent, accurate, and complete patient medical records in a pediatric hospital. Although the occurrence of incomplete medical records has been a known issue for the project hospital, the issue was further intensified following the implementation of the 10th revision of International Classification of Diseases (ICD-10) standard for documentation, which resulted in gaps in provider documentation that needed to be filled. Based on this, the researcher recommended a quality improvement project and worked with a multidisciplinary team from the hospital to develop an evidence-based documentation guideline that incorporated ICD-10 standard for documenting pediatric diagnoses. Using data generated from the guideline, an artificial intelligence (AI) was developed in the form of best practice advisory alerts to engage providers at the point of documentation as well as augment provider efforts. Rosswurm and Larrabee's conceptual framework and Kotter's 8-step change model was used to develop the guideline and design the project. A descriptive data analysis using sample T-test significance indicated that financial reimbursement decreased by 25%, while case denials increased by 28% after ICD-10 implementation. This project promotes positive social change by improving safety, quality, and accountability at the project hospital.

No description available.