A decision support system to facilitate management of patients with acute gastrointestinal bleeding

Objective: To develop a model to predict the bleeding source and identify the cohort amongst patients with acute gastrointestinal bleeding (GIB) who require urgent intervention, including endoscopy. Patients with acute GIB, an unpredictable event, are most commonly evaluated and managed by non-gastroenterologists. Rapid and consistently reliable risk stratification of patients with acute GIB for urgent endoscopy may potentially improve outcomes amongst such patients by targeting scarce health-care resources to those who need it the most. Design and methods: Using ICD-9 codes for acute GIB, 189 patients with acute GIB and all. available data variables required to develop and test models were identified from a hospital medical records database. Data on 122 patients was utilized for development of the model and on 67 patients utilized to perform comparative analysis of the models. Clinical data such as presenting signs and symptoms, demographic data, presence of co-morbidities, laboratory data and corresponding endoscopic diagnosis and outcomes were collected. Clinical data and endoscopic diagnosis collected for each patient was utilized to retrospectively ascertain optimal management for each patient. Clinical presentations and corresponding treatment was utilized as training examples. Eight mathematical models including artificial neural network (ANN), support vector machine (SVM), k-nearest neighbor, linear discriminant analysis (LDA), shrunken centroid (SC), random forest (RF), logistic regression, and boosting were trained and tested. The performance of these models was compared using standard statistical analysis and ROC curves. Results: Overall the random forest model best predicted the source, need for resuscitation, and disposition with accuracies of approximately 80% or higher (accuracy for endoscopy was greater than 75%). The area under ROC curve for RF was greater than 0.85, indicating excellent performance by the random forest model Conclusion: While most mathematical models are effective as a decision support system for evaluation and management of patients with acute GIB, in our testing, the RF model consistently demonstrated the best performance. Amongst patients presenting with acute GIB, mathematical models may facilitate the identification of the source of GIB, need for intervention and allow optimization of care and healthcare resource allocation; these however require further validation. (c) 2007 Elsevier B.V. All rights reserved.

Problems with accessibility to health services by persons with disabilities in Sao Paulo, Brazil

Purpose. To describe the occurrence of self-reported problems of accessibility to health services used by persons with disabilities in terms of social and health services variables. Methods. We performed a cross-sectional household survey designed to assess problems with accessibility to health services faced by persons with disabilities. We interviewed 333 persons in Sao Paulo city, in 2007. Variables related to the presence of accessibility problems, disabilities, gender, age, family head income, ethnicity, use of health services and others were analysed using frequencies, percentages, chi(2)-test, ANOVA and Poisson regression models. Results. 15.92% of the interviewed persons reported problems with accessibility to health services. Persons having multiple (prevalence ratios; PR = 2.91) or mobility disability (PR = 6.46) had more problems with accessibility than persons with hearing disability. Persons younger than 78 years old had more problems with accessibility; those who needed help to go to the health service (PR = 3.01) also. Conclusions. Persons with multiple or mobility disability, younger than 78 years, and those who needed help of others to go to the health service were more likely to have problems with accessibility to health services. This information could be one of the first steps to the management and/or planning of appropriate health services for persons with disabilities.

Barriers to interventions aimed at promoting the health of health care workers in Brazil

Objective. To search the literature for circumstances that impede injury and disease prevention and other activities intended to improve the health of the health care worker. Methods. The SciELO database was searched for articles published in 1967-2008. This was supplemented by a PubMed search for the period 1950-2008. The following key words were used to identify articles in English, Portuguese, and Spanish: work, health personnel, occupational, risks, diseases, ergonomics, work ability, quality of life, organization, accidents, work conditions, intervention, and administration. Articles on injury and disease prevention and occupational health in a health care setting in Latin America were selected, along with articles focused on health promotion in the health sector. Results. The following shortcomings were identified: activities lacked a sound theoretical foundation and were not integrated with the health services management; a failure to evaluate the effectiveness of the activity; health surveillance focused solely on a specific disease or injury; management not committed to the proposed activity; miscommunication; inability of workers to participate, or control the work environment; and, programs or efforts that were limited to changing the workers` behaviors. Conclusions. The literature shows that all the barriers identified by this study affect both the health care workers` health as well as their productivity.

Critical, issues and needs in management of primary immunodeficiency diseases in Latin America

Experts from six Latin American countries met to discuss critical issues and needs in the diagnosis and management of primary immunodeficiency diseases (PIDD). The diagnosis of PIDD is generally made following referral to an immunology centre located in a major city, but many paediatricians and general practitioners are not sufficiently trained to suspect PIDD in the first place. Access to laboratory testing is generally limited, and only some screening tests are typically covered by government health programmes. Specialised diagnostic tests are generally not reimbursed. Access to treatment varies by country reflecting differences in healthcare systems and reimbursement policies. An online PIDD Registry Programme for Latin America has been available since 2009, which will provide information about PIDD epidemiology in the region. Additional collaboration across countries appears feasible in at least two areas: a laboratory network to facilitate the diagnosis of PIDD, and educational programmes to improve PIDD awareness. In total, these collaborations should make it possible to advance the diagnosis and management of PIDD in Latin America. (C) 2010 SEICAP. Published by Elsevier Espana, S.L. All rights reserved.

Asthma management and asthma control in Sao Paulo, Brazil and Uppsala, Sweden: a questionnaire-based comparison

The Global Initiative Against Asthma (GINA) was developed to meet the global challenge of asthma. GINA has been adopted in most countries and comparison of asthma management in different parts of the world may be of help when assessing the global dissemination of the guideline. The overall goals in GINA include that asthma patients should be free of symptoms, acute asthma attacks and activity limitations. The aim of the present study was to compare asthma management and asthma control in Sao Paulo, Brazil and Uppsala, Sweden. Information was collected from asthmatics in Sao Paulo and Uppsala with a questionnaire. The questionnaire dealt with the following issues: symptoms, smoking, self-management, hospital visits, effect on school/work and medication. The Sao Paulo patients were more likely to have uncontrolled asthma (36% vs 13%, P < 0.001), having made emergency room visits (57% vs 29%, P < 0.001) and having lost days at school or work because of their asthma (46% vs 28%, P = 0.03) than the asthmatics from Uppsala. There were no difference in the use of inhaled corticosteroids, but the Brazilian patients were more likely to be using theophylline (18% vs 1%, P = 0.001) and less likely to be using long-acting beta-2 agonists (18% vs 37%, P < 0.001). We conclude that the level of asthma control was lower among the patients from Sao Paulo than Uppsala. Few of the patients in either city reached the goals set up by GINA. Improved asthma management may therefore lead to health-economic benefits in both locations. Please cite this paper as: Skorup P, Rizzo LV, Machado-Boman L and Janson C. Asthma management and asthma control in Sao Paulo, Brazil and Uppsala, Sweden: a questionnaire-based comparison. The Clinical Respiratory Journal 2009; 3: 22-28.