Using Machine Learning Classifiers to Assist Healthcare-Related Decisions: Classification of Electronic Patient Records

Surveillance Levels (SLs) are categories for medical patients (used in Brazil) that represent different types of medical recommendations. SLs are defined according to risk factors and the medical and developmental history of patients. Each SL is associated with specific educational and clinical measures. The objective of the present paper was to verify computer-aided, automatic assignment of SLs. The present paper proposes a computer-aided approach for automatic recommendation of SLs. The approach is based on the classification of information from patient electronic records. For this purpose, a software architecture composed of three layers was developed. The architecture is formed by a classification layer that includes a linguistic module and machine learning classification modules. The classification layer allows for the use of different classification methods, including the use of preprocessed, normalized language data drawn from the linguistic module. We report the verification and validation of the software architecture in a Brazilian pediatric healthcare institution. The results indicate that selection of attributes can have a great effect on the performance of the system. Nonetheless, our automatic recommendation of surveillance level can still benefit from improvements in processing procedures when the linguistic module is applied prior to classification. Results from our efforts can be applied to different types of medical systems. The results of systems supported by the framework presented in this paper may be used by healthcare and governmental institutions to improve healthcare services in terms of establishing preventive measures and alerting authorities about the possibility of an epidemic.

Electronic patient records: Achieving best practice in information transfer between hospital and community providers - an integration success story

Effective healthcare integration is underpinned by clinical information transfer that is timely, legible and relevant. The aim of this study was to describe and evaluate a method for best practice information exchange. This was achieved based on the generic Mater integration methodology. Using this model the Mater Health Services have increased effective community fax discharge from 34% in 1999 to 86% in 2002. These results were predicated on applied information technology excellence involving the development of the Mater Electronic Health Referral Summary and effective change management methodology, which included addressing issues around patient consent, engaging clinicians, provision of timely and appropriate education and training, executive leadership and commitment and adequate resourcing. The challenge in achieving best practice information transfer is not solely in the technology but also in implementing the change process and engaging clinicians. General practitioners valued the intervention highly. Hospital and community providers now have an inexpensive, effective product for critical information exchange in a timely and relevant manner, enhancing the quality and safety of patient care.

Mining free-text medical records for companion animal enteric syndrome surveillance.

Large amounts of animal health care data are present in veterinary electronic medical records (EMR) and they present an opportunity for companion animal disease surveillance. Veterinary patient records are largely in free-text without clinical coding or fixed vocabulary. Text-mining, a computer and information technology application, is needed to identify cases of interest and to add structure to the otherwise unstructured data. In this study EMR's were extracted from veterinary management programs of 12 participating veterinary practices and stored in a data warehouse. Using commercially available text-mining software (WordStat™), we developed a categorization dictionary that could be used to automatically classify and extract enteric syndrome cases from the warehoused electronic medical records. The diagnostic accuracy of the text-miner for retrieving cases of enteric syndrome was measured against human reviewers who independently categorized a random sample of 2500 cases as enteric syndrome positive or negative. Compared to the reviewers, the text-miner retrieved cases with enteric signs with a sensitivity of 87.6% (95%CI, 80.4-92.9%) and a specificity of 99.3% (95%CI, 98.9-99.6%). Automatic and accurate detection of enteric syndrome cases provides an opportunity for community surveillance of enteric pathogens in companion animals.

Leximancer concept mapping of patient case studies

Quantitative databases are limited to information identified as important by their creators, while databases containing natural language are limited by our ability to analyze large unstructured bodies of text. Leximancer is a tool that uses semantic mapping to develop concept maps from natural language. We have applied Leximancer to educational based pathology case notes to demonstrate how real patient records or databases of case studies could be analyzed to identify unique relationships. We then discuss how such analysis could be used to conduct quantitative analysis from databases such as the Coronary Heart Disease Database.

A importância da intervenção nutricional na redução do peso corpóreo em pacientes com síndrome do anticorpo antifosfolípide

OBJETIVO: Avaliar a eficácia da intervenção nutricional na redução do excesso de peso (EP), em pacientes com síndrome do anticorpo antifosfolípide (SAF). MÉTODO: Incluídos 40 pacientes, acima de 18 anos, com diagnóstico de SAF primária ou secundária, acompanhados no Serviço de Reumatologia do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo (HC-FMUSP) e recrutados no período de outubro de 2005 a fevereiro de 2006. Foram coletados dados demográficos e realizados a revisão de prontuários, a mensuração de peso e da altura, o cálculo do índice de massa corpórea (IMC) atual e a adequação da dieta pelo IMC ideal. Foi realizada avaliação na primeira consulta e após intervalo mínimo de dois meses. RESULTADOS: A média de idade dos pacientes foi de 41 anos, sendo 93% de mulheres. Os pacientes com SAF primária perfaziam 25%, com média de dois anos da doença. A trombose venosa ocorreu em 63%, arterial em 48% e manifestações obstétricas em 27%. Na primeira consulta, 68% apresentavam EP, 27% eram eutróficos e 5% estavam com baixo peso (BP). Após três meses de intervenção, os eutróficos mantiveram o peso e os de BP tornaram-se eutróficos, segundo o IMC. Interessantemente, entre os pacientes com EP (n = 27), 82% emagreceram, 14% engordaram e 4% se mantiveram. Especificamente, 11 pacientes apresentaram 1% a 3% de perda ponderal de peso, oito perderam de 4% a 7%, dois reduziram 8% a 9% e um reduziu 13,6% com o acompanhamento nutricional. CONCLUSÃO: Foi demonstrado no presente estudo que a intervenção nutricional conseguiu atingir metas para redução de peso, possibilitando diminuição no risco trombótico num curto período, sendo, portanto, uma modalidade terapêutica inicial e de eleição para corrigir o EP em pacientes com SAF.

Factors influencing wound healing after surgery for metastatic disease of the spine

Study Design, The study group consisted of 53 patients who underwent 75 operations for spine metastases. Patient and tumor demographic factors, preoperative nutritional status, and perioperative adjunctive therapy were retrospectively reviewed. Objective, To determine the risk factors for wound breakdown and infection in patients undergoing surgery for spinal metastases. Summary of Background Data. Spinal Fusion using spine implants may be associated with an infection rate of 5% or more. Surgery for spine metastases is associated with an infection rate of more than 10%. Factors other than the type of surgery performed may account for the greater infection rate. Methods. Data were obtained by reviewing patient records. Age, sex, and neurologic status of the patient; tumor type and site; and surgical details were noted. Adjunctive treatment with corticosteroids and radiotherapy was recorded, Nutritional status was evaluated by determining serum protein and serum albumin concentrations and by total lymphocyte count. Results. Wound breakdown and Infection occurred in 75 of 75 wounds. No patient or tumor demographic factors other than intraoperative blood loss (P < 0.1) were statistically associated with infection; The correlation between preoperative protein deficiency (P < 0.01) or perioperative corticosteroid administration (P < 0.10) and wound infection was significant. There was no statistical correlation between lymphocyte count or perioperative radiotherapy and wound infection. Conclusions, The results indicate that preoperative protein depletion and perioperative administration of corticosteroids are risk factors for wound infection in patients undergoing surgery for spine metastases, Perioperative correction of nutritional depletion and cessation of steroid therapy may reduce wound complications.

Quality of diabetes care and coronary heart disease absolute risk in patients with type 2 diabetes mellitus in Australian general practice

Objective: To examine the quality of diabetes care and prevention of cardiovascular disease (CVD) in Australian general practice patients with type 2 diabetes and to investigate its relationship with coronary heart disease absolute risk (CHDAR). Methods: A total of 3286 patient records were extracted from registers of patients with type 2 diabetes held by 16 divisions of general practice (250 practices) across Australia for the year 2002. CHDAR was estimated using the United Kingdom Prospective Diabetes Study algorithm with higher CHDAR set at a 10 year risk of >15%. Multivariate multilevel logistic regression investigated the association between CHDAR and diabetes care. Results: 47.9% of diabetic patient records had glycosylated haemoglobin (HbA1c) >7%, 87.6% had total cholesterol >= 4.0 mmol/l, and 73.8% had blood pressure (BP) >= 130/85 mm Hg. 57.6% of patients were at a higher CHDAR, 76.8% of whom were not on lipid modifying medication and 66.2% were not on antihypertensive medication. After adjusting for clustering at the general practice level and age, lipid modifying medication was negatively related to CHDAR (odds ratio (OR) 0.84) and total cholesterol. Antihypertensive medication was positively related to systolic BP but negatively related to CHDAR (OR 0.88). Referral to ophthalmologists/optometrists and attendance at other health professionals were not related to CHDAR. Conclusions: At the time of the study the diabetes and CVD preventive care in Australian general practice was suboptimal, even after a number of national initiatives. The Australian Pharmaceutical Benefits Scheme (PBS) guidelines need to be modified to improve CVD preventive care in patients with type 2 diabetes.

Clinicopathological evaluation of in vivo epidermal nuclear fluorescence

Nuclear fluorescence in keratinocytes is an occasional phenomenon, often present in autoimmune diseases, especially in connective-tissue disease (CTD); however, its clinical significance remains unclear. To investigate the profile of patients with positive nuclear staining on direct immunofluorescence (DIF) of skin samples. A retrospective analysis of 28 patient records from our immunodermatology laboratory was performed between May 2003 and June 2006. Inclusion criteria were the presence of autoantibodies (IgG, IgA or IgM) or complement (C3) binding keratinocyte nuclei on DIF. The most prevalent diseases related to the nuclear keratinocyte DIF staining were systemic lupus erythematosus (n = 9), mixed CTD (n = 3), overlap syndrome (n = 3), Sjogren`s syndrome (n = 1), and CREST (calcinosis, Raynaud`s phenomenon, oesophageal dysmotility, sclerodactyly and telangiectasia) syndrome (n = 1). Serum antinuclear antibody (ANA) was positive in 20 of 28 patients, with titres varying from 1 : 160 to 1 : 1280. Of the 20 patients with positive anti-nuclear antibodies (ANA), 17 were positive for anti-extractable nuclear antigen antibodies, 12 had anti-SSA/Ro, 11 had anti-SSB/La and 8 had anti-ribonucleoprotein. Eight patients were negative for ANA. Positive predictive value of in vivo ANA for systemic CTDs was 75%. The present data suggest that in vivo ANA evaluation is an additional and feasible auxiliary tool for diagnosing CTDs.

O financiamento por capitação ajustada pelo risco em contexto de integração vertical de cuidados de saúde : a utilização dos consumos com medicamentos como proxy da carga de doença em ambulatório

RESUMO - Introdução - Com o presente projecto de investigação pretendeu-se estudar o financiamento por capitação ajustado pelo risco em contexto de integração vertical de cuidados de saúde, recorrendo particularmente a informação sobre o consumo de medicamentos em ambulatório como proxy da carga de doença. No nosso país, factores como a expansão de estruturas de oferta verticalmente integradas, inadequação histórica da sua forma de pagamento e a recente possibilidade de dispor de informação sobre o consumo de medicamentos de ambulatório em bases de dados informatizadas são três fortes motivos para o desenvolvimento de conhecimento associado a esta temática. Metodologia - Este trabalho compreende duas fases principais: i) a adaptação e aplicação de um modelo de consumo de medicamentos que permite estimar a carga de doença em ambulatório (designado de PRx). Nesta fase foi necessário realizar um trabalho de selecção, estruturação e classificação do modelo. A sua aplicação envolveu a utilização de bases de dados informatizadas de consumos com medicamentos nos anos de 2007 e 2008 para a região de Saúde do Alentejo; ii) na segunda fase foram simulados três modelos de financiamento alternativos que foram propostos para financiar as ULS em Portugal. Particularmente foram analisadas as dimensões e variáveis de ajustamento pelo risco (índices de mortalidade, morbilidade e custos per capita), sua ponderação relativa e consequente impacto financeiro. Resultados - Com o desenvolvimento do modelo PRx estima-se que 36% dos residentes na região Alentejo têm pelo menos uma doença crónica, sendo a capacidade de estimação do modelo no que respeita aos consumos de medicamentos na ordem dos 0,45 (R2). Este modelo revelou constituir uma alternativa a fontes de informação tradicionais como são os casos de outros estudos internacionais ou o Inquérito Nacional de Saúde. A consideração dos valores do PRx para efeitos de financiamento per capita introduz alterações face a outros modelos propostos neste âmbito. Após a análise dos montantes de financiamento entre os cenários alternativos, obtendo os modelos 1 e 2 níveis de concordância por percentil mais próximos entre si comparativamente ao modelo 3, seleccionou-se o modelo 1 como o mais adequado para a nossa realidade. Conclusão - A aplicação do modelo PRx numa região de saúde permitiu concluir em função dos resultados alcançados, que já existe a possibilidade de estruturação e operacionalização de um modelo que permite estimar a carga de doença em ambulatório a partir de informação relativa ao seu perfil de consumo de medicamentos dos utentes. A utilização desta informação para efeitos de financiamento de organizações de saúde verticalmente integradas provoca uma variação no seu actual nível de financiamento. Entendendo este estudo como um ponto de partida onde apenas uma parte da presente temática ficará definida, outras questões estruturantes do actual sistema de financiamento não deverão também ser olvidadas neste contexto. ------- ABSTRACT - Introduction - The main goal of this study was the development of a risk adjustment model for financing integrated delivery systems (IDS) in Portugal. The recent improvement of patient records, mainly at primary care level, the historical inadequacy of payment models and the increasing number of IDS were three important factors that drove us to develop new approaches for risk adjustment in our country. Methods - The work was divided in two steps: the development of a pharmacy-based model in Portugal and the proposal of a risk adjustment model for financing IDS. In the first step an expert panel was specially formed to classify more than 33.000 codes included in Portuguese pharmacy national codes into 33 chronic conditions. The study included population of Alentejo Region in Portugal (N=441.550 patients) during 2007 and 2008. Using pharmacy data extracted from three databases: prescription, private pharmacies and hospital ambulatory pharmacies we estimated a regression model including Potential Years of Life Lost, Complexity, Severity and PRx information as dependent variables to assess total cost as the independent variable. This healthcare financing model was compared with other two models proposed for IDS. Results - The more prevalent chronic conditions are cardiovascular (34%), psychiatric disorders (10%) and diabetes (10%). These results are also consistent with the National Health Survey. Apparently the model presents some limitations in identifying patients with rheumatic conditions, since it underestimates prevalence and future drug expenditure. We obtained a R2 value of 0,45, which constitutes a good value comparing with the state of the art. After testing three scenarios we propose a model for financing IDS in Portugal. Conclusion - Drug information is a good alternative to diagnosis in determining morbidity level in a population basis through ambulatory care data. This model offers potential benefits to estimate chronic conditions and future drug costs in the Portuguese healthcare system. This information could be important to resource allocation decision process, especially concerning risk adjustment and healthcare financing.

Occurrence of 102 cases of paracoccidioidomycosis in 18 months in the Itaipu Lake region, Western Paraná

INTRODUCTION: The study investigated the incidence of disease and death events among patients with paracoccidioidomycosis who were residents in the Itaipu Lake region from 2008 to 2009. METHODS: A review of patient records was conducted at the Department of Tuberculosis of the Epidemiology Center of the City of Foz do Iguaçu, Paraná. RESULTS: The results identified 102 new cases of paracoccidioidomycosis in the period described, 72 men and 30 women, and 15 deaths were recorded during the study. CONCLUSIONS: It can be concluded that the Itaipu Lake region is an endemic region.

Evaluation of preventive measures for mother-to-child transmission of HIV in Aracaju, State of Sergipe, Brazil

INTRODUCTION: The main route of human immunodeficiency virus (HIV) infection in children is from mother to child. The preventive measures established for the Aids Clinical Trial Group protocol 076 (ACTG 076) significantly reduces HIV vertical transmission rates. This study aims to evaluate the implementation of the ACTG 076 protocol in the maternity units of State of Sergipe, Brazilian northeast. METHODS: This is a descriptive, retrospective study with a quantitative approach, with HIV positive women and children exposed, attending a Maternity reference for high-risk pregnancies. Data were obtained from patient records registered in the years 1994 to 2010. RESULTS: Amongst the 110 pregnant women and exposed newborns, the ACTG 076 protocol was fully utilized in only 31.8% of the participants. During the prenatal period, zidovudine (ZDV) was taken by 79.1% of the pregnant women. Only 49.1% of HIV seropositive patients used ZDV during delivery. Two (1.8%) children were considered infected and 50 (45.5%) do not have a conclusive diagnosis to date. CONCLUSIONS: There were significant deficiencies in the prevention of mother-to-child transmission of HIV, including lack of compliance with the three phases of the ACTG 076 protocol; inadequacies in prenatal care; inappropriate mode of delivery and lack of adequate follow up of exposed children.

Current epidemiological profile of Chagasic megaesophagus in Central Brazil

Introduction Chagasic megaesophagus (CM) is the most common digestive manifestation of Chagas disease in Brazil, and the State of Goiás is one of the most affected regions. In recent decades, the Hospital das Clínicas (HC)/Universidade Federal de Goiás (UFG) has been a reference center for the study and treatment of CM. The objective of this study was to characterize the current epidemiological profile of patients with CM observed at the HC of the UFG from 1998 to 2010. Methods In total, 939 patient records were analyzed, and age, gender, place of birth, serology, symptoms and radiological classification according to Rezende et al. were analyzed. Results The median patient age was 55 years. Male patients were more (54%) prevalent than female patients. The prevalence of younger patients (less than 31 years of age) was 4.2%, but 82.1% of the younger patients were from State of Bahia. Patients older than 40 years were the majority (85.5%). The radiological groups were distributed as follows: Group I (35.9%), Group II (32.9%), Group III (17%) and Group IV (14.2%). Conclusions Compared with previous studies by the same group in 1975, 1994 and 1995, the number of younger patients decreased, and the frequency curve has shifted to older patients.

Histoplasmosis in immunocompetent individuals living in an endemic area in the Brazilian Southeast

Introduction The distribution of infection by Histoplasma capsulatum in Brazil is heterogeneous, and the number of cases affecting immunocompetent individuals is relatively small. This study reports the epidemiological and clinical data regarding histoplasmosis in non-immunosuppressed individuals. Methods The study included only the immunocompetent patients with histoplasmosis who were diagnosed between 1970 and 2012 at a university hospital located in Ribeirão Preto, State of São Paulo, Brazil. Clinical and epidemiological data were collected retrospectively from the patient records. Results Of the 123 patients analyzed, 95 had an active disease that manifested in the different clinical forms of histoplasmosis. Men were the predominant gender, and most patients resided in the Northeast of the State of São Paulo and in the nearby municipalities of the State of Minas Gerais. The risk factors for acquiring histoplasmosis and prolonged contact in a rural environment were recorded in 43.9% and 82.9% of cases, respectively. Smoking, alcoholism, and comorbidity rates were high among the patients with the chronic pulmonary and subacute/chronic disseminated forms of histoplasmosis. Many patients achieved clinical cure spontaneously, but 58.9% required antifungals; the disease lethality rate was 5.3%. Conclusions Immunocompetent individuals manifested the diverse clinical forms of histoplasmosis over a period of 4 decades, revealing an additional endemic area of this fungal disease in the Brazilian Southeast.