Is it possible to estimate the number of patients with COVID-19 admitted to intensive care units and general wards using clinical and telemedicine data?

ABSTRACT Objective: To develop and validate predictive models to estimate the number of COVID-19 patients hospitalized in the intensive care units and general wards of a private not-for-profit hospital in São Paulo, Brazil. Methods: Two main models were developed. The first model calculated hospital occupation as the difference between predicted COVID-19 patient admissions, transfers between departments, and discharges, estimating admissions based on their weekly moving averages, segmented by general wards and intensive care units. Patient discharge predictions were based on a length of stay predictive model, assessing the clinical characteristics of patients hospitalized with COVID-19, including age group and usage of mechanical ventilation devices. The second model estimated hospital occupation based on the correlation with the number of telemedicine visits by patients diagnosed with COVID-19, utilizing correlational analysis to define the lag that maximized the correlation between the studied series. Both models were monitored for 365 days, from May 20th, 2021, to May 20th, 2022. Results: The first model predicted the number of hospitalized patients by department within an interval of up to 14 days. The second model estimated the total number of hospitalized patients for the following 8 days, considering calls attended by Hospital Israelita Albert Einstein's telemedicine department. Considering the average daily predicted values for the intensive care unit and general ward across a forecast horizon of 8 days, as limited by the second model, the first and second models obtained R² values of 0.900 and 0.996, respectively and mean absolute errors of 8.885 and 2.524 beds, respectively. The performances of both models were monitored using the mean error, mean absolute error, and root mean squared error as a function of the forecast horizon in days. Conclusion: The model based on telemedicine use was the most accurate in the current analysis and was used to estimate COVID-19 hospital occupancy 8 days in advance, validating predictions of this nature in similar clinical contexts. The results encourage the expansion of this method to other pathologies, aiming to guarantee the standards of hospital care and conscious consumption of resources.

Validación y Comparación de un Nuevo Método para Predecir el Ancho Mesiodistal de Caninos y Premolares no Erupcionados en una Población Chilena de la Región de Valparaíso, Chile

Uno de los principales problemas durante la dentición mixta es la determinación de la futura discrepancia entre tamaño dentario y el espacio disponible. Para predecir el ancho mesiodistal de los dientes permanentes no erupcionados se han introducido diferentes métodos de análisis. Objetivo: El propósito de este estudio fue comparar el método Tanaka-Johnston con una nueva ecuación de regresión para predecir el ancho mesiodistal de caninos y premolares permanentes no erupcionados en una población de la región de Valparaíso, Chile. Material y método: Este estudio fue realizado en la Facultad de Odontología de la Universidad de Valparaíso, desde octubre de 2022 a junio de 2023 (8 meses), la muestra estuvo compuesta por 202 modelos de estudio del departamento de ortodoncia (91 hombres y 111 mujeres) en el rango de edad de 11 -20 años. Resultados: Se demostró que el método elaborado por Lara-Sandoval presenta mayor fiabilidad respecto a las medidas mesiodistales reales de los pacientes (ICC 0,773 para maxilar y 0,762 para mandíbula), en comparación con el método de Tanaka-Johnston (ICC 0,665 para maxilar y 0,623 para mandíbula). No existen diferencias significativas entre los valores reales y el método de Lara-Sandoval. Conclusión: El método de Lara-Sandoval es mejor que el propuesto por Tanaka-Johnston para determinar el ancho mesiodistal de caninos y premolares para esta muestra. Es necesario validar este método en otras regiones del país para ser utilizado con mayor seguridad que el ya existente como método estándar nacional.

One of the main orthodontic problems in mixed dentition is the determination of future tooth and size arch discrepancy. In order to predict the mesiodistal widths of unerupted permanent teeth different methods of analyses have been introduced. The aim of this study is to compare the Tanaka-Johnston analysis with a new regressive equation to predict the mesiodistal width of unerupted permanent canines and premolars in a Chilean population sample, from Valparaíso region. This study was conducted at the Universidad de Valparaíso Dental Faculty, from october 2022 to june 2023 (8 months), and the sample comprised historical dental casts from 202 patients (91 boys and 111 girls) in the age range of 11-20 from the orthodontics department. All the patients are from the Valparaíso region, Chile. The results show that the predictions of the new regressive equation method are closer to the actual mesiodistal measurements of the patients (ICC 0,773 for maxilla and 0,762 for mandible), compared to the Tanaka- Johnston method (ICC 0,665 for maxilla and 0,623 for mandible). There are no significant differences between the real values and the Lara-Sandoval method. Lara-Sandoval method is better than the one proposed by Tanaka-Johnston to determine the mesiodistal width of canines and premolars in this sample population. It is necessary to validate this method in other regions of the country to be used with greater security than the ones that already exists as a national standard method.

Analysis of Watson for oncology and clinicians� treatment recommendations for patients with breast cancer in Korea: A single center experience

Background: Various clinical applications have been attempted using artificial intelligence (AI) clinical decision support system (CDSS), and it has become a starting point for personalized cancer treatment. We aimed to identify the degree of agreement between the AI?CDSS, Watson for Oncology (WFO), and the clinician in treatment recommendations for Korean breast cancer patients and to provide guidelines for future improvement. Methods: One hundred and eighty?three breast cancer patients who underwent treatment at the Pusan National University Hospital between January 1, 2016 and May 31, 2017 were enrolled in this study. The concordance between WFO抯 and clinicians� treatment recommendations were examined according to various factors. Results: WFO gave the same treatment option recommendations as clinicians in 74 (40.4%) of the cases. According to the logistic regression, the difference in recommendation concordance between stage I and stage III was statistically significant (P = 0.004), and there was no difference among other factors. Conclusion: The concordance of treatment recommendations was low overall. However, this is largely attributable to the differences of medical insurance system and healthcare environment between the United States and Korea. In the future, region?specific features should be considered or reflected during the development of AI?CDSS.

Study on safety performance and condition-suggestion accuracy of the symptom assessment mobile applications / 中华全科医师杂志

Objective:To compare the breadth of condition coverage, accuracy of suggested conditions and appropriateness of urgency advice of the 8 symptom assessment mobile applications (APPs) available on the Chinese market.Methods:The APPs were assessed using 200 primary care vignettes and were measured against the vignettes′ standard. The primary outcome measures were proportion of conditions covered by an APP, proportion of vignettes with the correct primary diagnosis,and proportion of safe urgency advice.Results:For APPs assessed,condition-coverage was from 29.0%(58/200)to 99.5%(199/200), top-3 suggestion accuracy was from 8.5%(17/200) to 61.5%(123/200), the proportion of safe urgency advice was from 84.8%(167/197) to 99.5% (198/199).Conclusions:The APPs showed a wide range of coverage, safety performance and condition-suggestion accuracy. Symptom assessment APPs with good performance could be used by general practitioners as supporting tools. However, even symptom assessment APPs with excellent performance need to be further assessed in a real clinical environment.

Application of Mengchao Liver Disease-Brain System version 2.0 in artificial intelligence-assisted clinical diagnosis and treatment： A preliminary study / 临床肝胆病杂志

ObjectiveTo investigate the application of Mengchao Liver Disease-Brain System version 2.0 in clinical diagnosis and treatment. MethodsThis study was conducted among 160 patients who were admitted to the internal medicine and surgical departments from June 9 to 21， 2021， and their data were automatically captured by the intelligent information system of Southeast Big Data Institute of Hepatobiliary Health， Mengchao Hepatobiliary Hospital of Fujian Medical University. The completeness and accuracy of Mengchao Liver Disease-Brain System version 2.0 were evaluated based on the intelligent diagnostic tools such as auxiliary diagnosis of chronic hepatitis B， interpretation of liver fibrosis， staging model of chronic hepatitis B， auxiliary diagnosis of liver cirrhosis， auxiliary staining of liver cirrhosis， auxiliary diagnosis of primary liver cancer， BCLC stage of primary liver cancer， Chinese staging of primary liver cancer， Child-Pugh score， and APRI score. ResultsAll auxiliary diagnostic tools had a complete rate of 94.17% in terms of the extraction of correct key dimensions within the test period. The artificial intelligence report had a structured accuracy of 97.55% in capturing data and an accuracy rate of 91.61% in text processing. ConclusionMengchao Liver Disease-Brain System version 2.0 provides an innovative mode for the construction of big data platform in medical specialties and has a high accuracy as an auxiliary diagnostic tool in clinical diagnosis and treatment.

Software para apoio à decisão logística no processo de captação e transplante de órgãos / Software de apoyo a la decisión logística en el proceso de obtención y trasplante de órganos / Software to support logistic decision in the process of organ acquisition and transplantation

Objetivo: desenvolver e avaliar um software para apoio à tomada de decisão dos profissionais da central de transplantes nas definições logísticas envolvidas no processo de captação e distribuição de órgãos para transplante. Método: estudo de produção tecnológica aplicada, sustentado pelo método Design Science Research Methodology. Participaram da etapa de avaliação da usabilidade dez enfermeiros da Central de Transplantes de Santa Catarina. A coleta de dados ocorreu de 1 a 20 de julho de 2021 por meio do questionário System Usability Scale. Resultados: o software utilizou linguagem JavaScript com ReactJS e PHP com Laravel, para o banco de dados PostgreSQL. A avaliação obteve escore médio de 98,25, sendo sua usabilidade considerada como melhor alcançável. Conclusão: o software demonstrou ser adequado e funcional, com fácil manuseio, reunindo informações integradas e objetivas. Representa um avanço na área, propondo uma inovação tecnológica para a gestão e apoio às decisões logísticas envolvidas no processo de captação e transplante de órgãos.

Objetivo: desenvolver e avaliar um software para apoio à tomada de decisão dos profissionais da central de transplantes nas definições logísticas envolvidas no processo de captação e distribuição de órgãos para transplante. Método: estudo de produção tecnológica aplicada, sustentado pelo método Design Science Research Methodology. Participaram da etapa de avaliação da usabilidade dez enfermeiros da Central de Transplantes de Santa Catarina. A coleta de dados ocorreu de 1 a 20 de julho de 2021 por meio do questionário System Usability Scale. Resultados: o software utilizou linguagem JavaScript com ReactJS e PHP com Laravel, para o banco de dados PostgreSQL. A avaliação obteve escore médio de 98,25, sendo sua usabilidade considerada como melhor alcançável. Conclusión: o software demonstrou ser adequado e funcional, com fácil manuseio, reunindo informações integradas e objetivas. Representa um avanço na área, propondo uma inovação tecnológica para a gestão e apoio às decisões logísticas envolvidas no processo de captação e transplante de órgãos.

Objective: to develop and evaluate a software to support the decision-making of transplant center professionals in the logistic definitions involved in the process of organ procurement and distribution for transplantation. Method: applied technological production study, supported by the Design Science Research Methodology method. Ten nurses from the Transplant Center of Santa Catarina participated in the usability evaluation stage. Data collection took place from 1 to 20 July 2021 through the System Usability Scale questionnaire. Results: the software used JavaScript language with ReactJS and PHP with Laravel, for the PostgreSQL database. The evaluation obtained a mean score of 98.25, and its usability is considered as best achievable. Conclusion: the software proved to be adequate and functional, with easy handling, gathering integrated and objective information. It represents a breakthrough in the area, proposing a technological innovation for the management and support to the logistic decisions involved in the process of organ procurement and transplantation.

Clinical decision support systems for diabetic foot ulcers: a scoping review / Sistemas de apoyo a decisiones clínicas para las úlceras del pie diabético: una revisión de alcance / Sistemas de apoio à decisão clínica em úlceras de pé diabético: revisão de escopo

ABSTRACT Objective: Map the scientific evidence on the use of clinical decision support systems in diabetic foot care. Method: A scoping review based on the JBI Manual for Evidence Synthesis and registered on the Open Science Framework platform. Searches were carried out in primary and secondary sources on prototypes and computerized tools aimed at assisting patients with diabetic foot or at risk of having it, published in any language or period, in eleven databases and grey literature. Results: A total of 710 studies were identified and, following the eligibility criteria, 23 were selected, which portrayed the use of decision support systems in diabetic foot screening, predicting the risk of ulcers and amputations, classifying the stage of severity, deciding on the treatment plan, and evaluating the effectiveness of interventions, by processing data relating to clinical and sociodemographic information. Conclusion: Expert systems stand out for their satisfactory results, with high precision and sensitivity when it comes to guiding and qualifying the decision-making process in diabetic foot prevention and care.

RESUMEN Objetivo: Mapeo de la evidencia científica sobre el uso de sistemas de apoyo a la toma de decisiones clínicas en el cuidado del pie diabético. Método: Revisión de alcance basada en el Manual de Síntesis de la Evidencia del JBI y registrada en la plataforma Open Science Framework. Se realizaron búsquedas en fuentes primarias y secundarias sobre prototipos y herramientas informatizadas dirigidas a la asistencia de pacientes con pie diabético o en riesgo de padecerlo, publicadas en cualquier idioma o periodo, en once bases de datos y literatura gris. Resultados: Se identificaron 710 estudios y, tras cumplir los criterios de elegibilidad, se seleccionaron 23, que retrataban el uso de sistemas de apoyo a la toma de decisiones en el diagnóstico del pie diabético, la predicción del riesgo de úlceras y amputaciones, la clasificación del estadio de gravedad, la decisión sobre el plan de tratamiento y la evaluación de la eficacia de las intervenciones, mediante el procesamiento de datos relativos a la información clínica y sociodemográfica. Conclusión: Los sistemas expertos destacan por sus resultados satisfactorios, con gran precisión y sensibilidad a la hora de orientar y cualificar el proceso de toma de decisiones en la prevención y el cuidado del pie diabético.

RESUMO Objetivo: Mapear as evidências científicas sobre uso de Sistemas de Apoio à Decisão Clínica no pé diabético. Método: Revisão de escopo fundamentada no JBI Manual for Evidence Synthesis e registrada na plataforma Open Science Framework. Realizaram-se buscas, em fontes primárias e secundárias, sobre protótipos e ferramentas informatizadas direcionadas à assistência ao paciente com pé diabético ou em risco de tê-lo, publicados em qualquer idioma ou período, em onze bases de dados e literatura cinza. Resultados: Foram identificados 710 estudos e, após critérios de elegibilidade, foram selecionados 23 que retratam o uso de sistemas de apoio à decisão no rastreio do pé diabético, predição do risco de úlceras e amputações, classificação do estágio de gravidade, decisão quanto ao plano de tratamento e avaliação da efetividade das intervenções, por meio do processamento de dados referentes a informações clínicas e sociodemográficas. Conclusão: Os sistemas especialistas destacam-se por resultados satisfatórios, com alta precisão e sensibilidade no que tange à orientação e qualificação do processo de tomada de decisão na prevenção e no cuidado ao pé diabético.

Sistemas de apoio à decisão clínica e desfecho clínico na Atenção Primária à Saúde: uma revisão sistemática / Clinical decision support systems and clinical outcomes in Primary Health Care: a systematic review

Esta revisão sistemática aborda o uso de Sistemas de Suporte à Decisão Clínica (SADC) nos atendimentos realizados na Atenção Primária à Saúde (APS), identificando relações existentes entre o uso dos sistemas e os desfechos clínicos. Foram selecionados trabalhos, estudos em português e inglês, sem restrição ao cenário brasileiro, encontrados em diferentes bases de dados. Os resultados demonstram que os SADC ainda se encontram em estágio de desenvolvimento e refinamento, com aplicação ainda incipiente nas mais diversas patologias e condições clínicas. São raros os ensaios clínicos que tracem os desfechos clínicos primários, levando ao acúmulo de dados apenas sobre desfechos secundários ou compostos, dificultando a avaliação dos sistemas. Há indicativos de relativa eficiência no uso dos SADC para situações de diagnóstico e prevenção, com eficiência limitada na fase de tratamento. Finalmente, não existem dados suficientes para afirmar se os SADC geram desfechos clínicos primários mais favoráveis ou desfavoráveis na APS.

This systematic review addresses the use of Clinical Decision Support Systems (CDSS) in Primary Health Care (PHC), identifying relationships between the use of the Systems and clinical outcomes. The research employed selected studies in Portuguese and English, with no restriction to the Brazilian scenario, found in different databases. Results demonstrate that CDSS are still in the development and refinement stage, and their application is still incipient for the most diverse pathologies and clinical conditions. Clinical trials that trace the primary clinical outcomes are rare, leading to the accumulation of data only on secondary or compound outcomes, making it difficult to evaluate the systems. There are indications of relative efficiency in the use of CDSS for diagnosis and prevention situations, with limited efficiency in the treatment phase. Finally, there is insufficient data to establish whether CDSS generates more favorable or unfavorable primary clinical outcomes in PHC.

How an electronic prescription tool enables better prescription quality for patients

Objective: Medication-related errors in patients are among the leading causes of preventable health damage and harm worldwide. In the United States, these errors cause at least one death a day and damage approximately 1.3 million people annually. According to the World Health Organization, the global expenditure on medication-related errors is estimated to be U$ 42 billion per year. In Brazil, the rate of potential drug interactions varies between 28% and 63.6% for primary care patients. The prevalence of drug interactions has increased following an aging population, increased chronic conditions, combined use of different drugs, and increased prescription drugs per patient. Methods: The data used for this study were obtained through the database from Nexodata do Brasil S.A a private health technology company with an electronic prescription system and a data intelligence area. Results: 65,867 electronic prescriptions were evaluated during 2019. Of these, 4,828 prescriptions had an average of 2.5 interactions. These interactive prescriptions were generated by 197 different doctors, totaling 24.5 prescriptions with interaction per doctor over 12 months. A total of 12,005 interactions were identified, 15.6% classified as mild, 70.9% as moderate, and 13.5% as severe. Conclusion: By implementing an electronic prescription tool, a reduction of 32.9% in the number of prescriptions with drug interaction was observed.

Effects of patient participation decision aid on decision conflict and risk perception of patients undergoing cardiac valve replacement / 中国实用护理杂志

Objective:To explore the effects of application of the patient participation decision aid in patients undergoing cardiac valve replacement.Methods:A total of 38 patients hospitalized for cardiac valve replacement from June to November, 2020 in the First Affiliated Hospital of Nanjing Medical University, were enrolled as the control group, and 38 patients admitted from January to May, 2020, were recruited as the experimental group by convenient sampling method. The control group was given routine nursing intervention, while the experimental group was carried out patient participation decision aid assistant program on the basis of routine nursing. The effects of intervention were assessed by Decision Conflict Sale (DCS), Preparation for Decision Making (PrepDM) and Risk Perception Questionnaire for patients with chronic diseases at admission and one day before operation.Results:There was no significant difference in the scores of DCS, PrepDM and Risk Perception Questionnaire for patients with chronic diseases at admission between the two groups( P>0.05). At one day before operation, the information& values scores dimension, support and valid decision dimension scores, decision uncertainty dimension scores and total scores of DCS were (12.37 ± 4.11) , (12.50 ± 4.65) , (3.74 ± 1.17), (28.61 ± 7.07) points in the experimental group, lower than those in the control group (15.11 ± 3.62) , (17.84 ± 4.25), (4.37 ± 1.30), (37.32 ± 6.57) points, the differences were statistically significant between the two groups( t values were 2.22-5.56, all P<0.05). At one day before operation, the score of PrepDM was (73.58 ± 5.32) points in the experimental group, higher than that in the control group (67.82 ± 4.89) points, the difference was statistically significant ( t=4.92, P<0.05). At one day before operation, the scores of economic risk dimension, physical and medical risk dimension, psychosocial risk dimension scores and total scores of Risk Perception Questionnaire for patients with chronic diseases were (11.45 ± 3.94), (12.39 ± 3.64), (7.21 ± 4.09), (31.05 ± 6.11) points in the experimental group, lower than those in the control group (13.50 ± 3.73), (15.82 ± 5.18), (9.32 ± 3.59), (38.63 ± 7.27) points, the differences were statistically significant ( t values were 2.33-4.92, all P<0.05). Conclusions:Patient participation decision aid program can effectively decrease decision conflict, promote preparation for decision making and alleviate risk perception of patients undergoing cardiac valve replacement.

O uso da inteligência artificial como ferramenta de apoio à gestão das ações em saúde na Secretaria de Estado da Saúde de Goiás / The use of artificial intelligence as a tool to support the management of health actions in the Goiás State Health Department

Introdução: A área de Tecnologia da Informação da Secretaria de Estado da Saúde de Goiás com o intuito de desenvolver soluções tecnológicas para apoio à tomada de decisão iniciou em 2010 o projeto na área de Business Intelligence que engloba, dentre outras ações, a construção de um data warehouse que agrega dados dos sistemas transacionais em conjunto com a elaboração de painéis de monitoramento que permitem uma visão integrada do todo para subsidiar a alta gestão e áreas técnicas da Secretaria de Estado da Saúde de Goiás (SES-GO), garantindo assim, decisões mais assertivas e a elaboração de políticas públicas mais pertinentes e fundamentadas. Este projeto evoluiu e atualmente as áreas de Ciência de Dados e Big Data também compõem área de analytics da SES-GO. Objetivo: o principal objetivo deste artigo é apresentar a plataforma FLINK que foi desenvolvida vislumbrando a democratização de dados e indicadores de saúde da Secretaria de Estado da Saúde de Goiás e compartilhar o processo de construção e implantação da mesma. Metodologia: Trata-se de um estudo de abordagem qualitativa, do tipo relato de experiência, no qual os autores relatam a construção e funcionalidades da plataforma FLINK. Considerações finais: A plataforma FLINK representa a efetivação da transparência da gestão pública no estado de Goiás em dados de saúde. A versatilidade e a flexibilidade do mesmo possibilita a melhoria das ações de saúde e beneficia todos os usuários do SUS

Introduction: The servers in the Business Intelligence (BI) area of the State Department of Health of Goiás, in order to develop technological solutions to support decision-making, started in 2010 to build a data warehouse that would aggregate data from traditional systems, and of monitoring panels, allowing a more integrated view of the whole, to support the senior management and technical areas of State Department of Health of Goiás (SES-GO), thus ensuring more assertive decisions and the development of more relevant and grounded public policies. Objective: The main objective of this article is to present the flink plataform that was developed with a view to the democratization of data and health status indicators of the Goiás state health department and to share the process of construction and implementation of the same. Method: This is a study with a qualitative approach, of the experience report type, in which the authors report the construction and functionalities of the FLINK plataform. Final considerations: FLINK represents the effectiveness of transparency in public management in the state of Goiás. Its versatility and flexibility enables the improvement of health actions and benefits all SUS users

Evidências na COVID-19: atenção primária, telessaúde e atenção hospitalar / Evidence in COVID-19: primary care, telehealth and hospital care

Introdução: A pandemia da COVID-19 representa um grande desafio para os sistemas de saúde de todo o mundo, com demandas sem precedentes em todos os níveis de atenção à saúde. Na atenção primária, além do cuidado dos indivíduos com sintomas gripais, a descontinuidade do acompanhamento de pacientes com doenças crônicas é causa de grande preocupação. Na atenção hospitalar, a necessidade imediata de assistir pacientes com formas graves e críticas da COVID-19, exigiu ampliação de leitos convencionais e de cuidados intensivos, recursos tecnológicos para suporte avançado de vida e capacitação em tempo recorde, à medida que avançavam os conhecimentos sobre a doença. Objetivos: O presente trabalho possui dois objetivos principais. Objetivo 1: Avaliar o impacto da pandemia da COVID-19 em uma coorte de pessoas com hipertensão e diabetes na atenção primária, além de desenvolver e implementar uma solução digital para melhorar o monitoramento no domicílio. Objetivo 2: Avaliar se o escore de risco ABC2-SPH é capaz de predizer a necessidade de ventilação mecânica em pacientes com COVID-19 e comparar seu desempenho ao de outros escores desenvolvidos para predizer ventilação mecânica, mortalidade e outros desfechos, inclusive em pacientes não COVID. Métodos: Para alcançar o objetivo 1: Foi desenvolvido um estudo multimetodológico. Uma avaliação quasi-experimental analisou o impacto da pandemia na frequência de consultas e controle de pacientes com hipertensão e diabetes em 34 unidades básicas de saúde (UBS) em 10 municípios do Vale do Mucuri, Minas Gerais, entre junho de 2017 e dezembro de 2020, considerando como ponto de corte o dia 14 de março de 2020, quando as medidas de restrição social tiveram início. Em seguida, um estudo de factibilidade desenvolveu um aplicativo com um sistema de apoio à decisão (SAD) para agentes comunitários de saúde (ACS) para identificar, durante a visita domiciliar, pacientes com hipertensão arterial (HAS) e/ou diabetes mellitus (DM) não controlados e referenciá-los para atendimento presencial na UBS. Um painel de especialistas avaliou a viabilidade, usabilidade e utilidade do aplicativo através de questionário específico. Para alcançar o objetivo 2: Estudo de coorte retrospectiva, derivado do estudo originalmente chamado "Avaliação do perfil laboratorial, radiológico e sintomatológico de pacientes infectados com o novo coronavírus 2019 (SARS-CoV-2) em hospitais do estado de Minas Gerais" que se tornou um estudo multicêntrico, realizado em 31 hospitais, em 17 cidades de cinco estados. O estudo incluiu pacientes que foram internados em dois períodos: março a setembro de 2020 e março a dezembro de 2021, com diagnóstico de COVID-19 confirmado. Neste estudo, o escore de risco ABC2-SPH, desenvolvido para predição de mortalidade intra-hospitalar por COVID-19, foi avaliado quanto à possibilidade de predizer a necessidade de ventilação mecânica e comparado a outros escores: CALL, PREDI-CO, SUM, STSS, COVID_IRS_NLR, CURB-65, SOFA e 4C Mortality Score. Resultados: Objetivo 1: Dos 5070 pacientes, 4810 (94,9%) tinham HAS, 1371 (27,4%) tinham DM e 1111 (23,1%) tinham as duas doenças. Houve redução significativa no número de consultas semanais quando comparados o período antes e depois do início das medidas de restrição social (107 [IQR 60,0, 153,0] vs 20,0 [IQR 7,0, 29,0], respectivamente, p <0,001. Apenas 15,2% de todos os pacientes retornaram para consulta durante a pandemia. Indivíduos com HAS que retornaram tinham níveis de pressão arterial sistólica (120,0 [IQR 120,0-140,0]) e diastólica (80,0 [IQR 80,0-80,0]) menores do que os níveis apresentados antes da pandemia por aqueles que não retornaram (130,0 [IQR 120,0-140,0] e 80,0 [IQR 80,0-90,0]), p<0,001. Além disso, aqueles que retornaram tiveram uma proporção maior de HAS controlada (64,3% vs 52,8%). Para o DM, não houve diferenças nos níveis de glicohemoglobina. Em relação ao SAD, os especialistas concordaram que os ACS podem incorporá-lo facilmente em suas rotinas e o aplicativo pode identificar pacientes em risco e melhorar o tratamento. Objetivo 2: Ao longo do estudo, foram incluídos 4.831 pacientes, com idade mediana de 59,0 (IQR 48,0, 70,0) anos e 46,3% do sexo feminino. Destes, 34,2% necessitaram de tratamento intensivo, 26,6% necessitaram de VMI e 18,7% faleceram. Os pacientes que necessitaram de VMI apresentaram maior prevalência de HAS, DM, obesidade e mortalidade quando comparados aos que não necessitaram (64,3% vs 2,3%, p<0,001). Com base no conjunto de dados imputados, o ABC2-SPH apresentou AUROC 0,677 (IC 95% 0,681-0,694). Considerando apenas os casos completos, a AUROC foi de 0,70 (IC 95% 0,68-0,72), tendo o melhor desempenho entre os escores com amostras maiores de casos completos. No geral, as discriminações dos escores foram de regular a ruim. O SOFA Score teve a maior sensibilidade, 0,84 (IC 95% 0,81-0,86). O escore SOFA teve a maior sensibilidade 0,84 (IC 95% 0,81-0,86). Conclusões: A pandemia da COVID-19 causou uma queda significativa no número de consultas de pacientes com HAS e DM na atenção primária. Um SAD para ACS mostrou-se viável e útil para identificar pacientes descontrolados em casa. O escore ABC2-SPH demonstrou melhor desempenho do que os outros escores, mas não com precisão suficiente para prever com segurança a necessidade de ventilação mecânica em pacientes hospitalizados com COVID-19. Novos estudos devem ser realizados para desenvolver escores fáceis de aplicar que tenham melhor calibração e discriminação para auxiliar na difícil decisão de instituir o suporte ventilatório avançado.

Introduction: The COVID-19 pandemic represents a major challenge for health systems around the world, with unprecedented demands on all levels of health care. In primary care, in addition to the care of individuals with flu-like symptoms, the discontinuity of monitoring patients with chronic diseases is a cause of great concern. In hospital care, the immediate need to assist patients with severe and critical forms of COVID-19 required the expansion of conventional and intensive care beds, technological resources for advanced life support and training in record time, as knowledge about the disease evolved. Objectives: The present study has two main objectives. Objective 1: To assess the impact of the COVID-19 pandemic on a cohort of people with hypertension and diabetes in primary care, and to develop and implement a digital solution to improve home monitoring. Objective 2: To assess whether the ABC2-SPH risk score can predict the need for invasive mechanical ventilation (IMV) in patients with COVID-19 and to compare its performance with other scores developed to predict IMV, mortality and other outcomes, including in non-COVID patients. Methods: To achieve objective 1: A multi-methodological study was developed. A quasi-experimental evaluation analyzed the impact of the pandemic on the frequency of consultations and control of patients with hypertension and diabetes in 34 primary health care centers (PHCC) in 10 municipalities in Mucuri Valley (Vale do Mucuri), Minas Gerais, between June 2017 and December 2020, considering March 14, 2020, as the cut-off point, when social restriction measures began. Then, a feasibility study developed an application with a decision support system (DSS) for community health workers (CHW) to identify, during home visits, patients with uncontrolled hypertension and/or diabetes and refer them for in person consultation at the PHCC. A panel of experts evaluated the app's feasibility, usability and usefulness through a specific questionnaire. To achieve objective 2: Retrospective cohort study, derived from the study originally called "Evaluation of the laboratory, radiological and symptomatologic profile of patients infected with the new coronavirus 2019 (SARS-CoV-2) in hospitals in the state of Minas Gerais", which became a multicenter study, carried out in 31 hospitals, in 17 cities in five states The study included patients who were hospitalized in two periods: March to September 2020 and March to December 2021, with a confirmed diagnosis of COVID-19 . In this study, the ABC2-SPH risk score, developed to predict in-hospital mortality from COVID-19, was evaluated for the possibility of predicting the need for mechanical ventilation and compared to other scores: CALL, PREDI-CO, SUM, STSS, COVID_IRS_NLR, CURB-65, SOFA and 4C Mortality Score Results: Objective 1: Of 5070 patients, 4810 (94.9%) had hypertension, 1371 (27.4%) had DM, and 1111 (23.1%) had both diseases. There was a significant reduction in the weekly number of consultations (107, IQR 60.0-153.0 before vs 20.0, IQR 7.0-29.0) after social restriction; P<.001. Only 15.2% (772/5070) of all patients returned for a consultation during the pandemic. Individuals with hypertension had lower systolic (120.0, IQR 120.0-140.0 mm Hg) and diastolic (80.0, IQR 80.0-80.0 mmHg) blood pressure than those who did not return (130.0, IQR 120.0-140.0 mm Hg) and (80.0, IQR 80.0-90.0 mm Hg; P<.001). Also, those who returned had a higher proportion of controlled hypertension (64.3% vs 52.8%). For DM, there were no differences in glycohemoglobin levels. Concerning the DSS, the experts agreed that the CHWs can easily incorporate it into their routines and the app can identify patients at risk and improve treatment. Objective 2: Throughout the study, 4831 patients were included, median age 59.0 (IQR 48.0, 70.0) years-old and 46.3% female. Of those, 34.2 % required intensive care treatment, 26.6% required IMV and 18.7% died. Patients who required IMV had higher prevalence of hypertension, diabetes, obesity, and mortality when compared to those who did not require it (64.3% vs 2.3%, p<0.001). Based on the imputed dataset, the ABC2-SPH AUROC was 0.677 (95% CI 0.681-0.694). Considering only complete cases, the AUROC was 0.70, having the best performance among scores that had larger samples of complete cases. When the data were imputed, was 0.67. Considering only complete cases, the AUROC was 0.70 (95% CI 0.68-0.72, having the best performance among scores that had larger samples of complete cases. Overall, the score discriminations ranged from poor to fair. The SOFA Score had the highest sensitivity, 0.84 (95% CI 0.81-0.86). Conclusions: The COVID-19 pandemic caused a significant drop in the number of consultations of patients with hypertension and diabetes in primary care. An SSD for CHW proved to be feasible and useful for identifying uncontrolled patients at home. ABC2-SPH demonstrated better performance than the other scores, but not accurately enough to reliably predict the need for IMV COVID-19 hospitalized patients. Research should continue to develop easy-to-use scores with better calibration and discrimination, given the importance of assisting clinicians in decision-making when initiating advanced ventilatory support.

Sensibilidade e especificidade de sistemas open access para detecção de interações medicamentosas potenciais / Sensitivity and specificity of open access systems to detect potential drug-drug interactions

Patient safety.Estudo transversal. Objetivo: avaliar a sensibilidade e especificidade de sistemas de rastreamento de acesso aberto para interações medicamentosas potenciais (IMp) em comparação com o DRUG-REAX® system e analisar o impacto clínico potencial das IMp de gravidades "Contraindicada" e "Maior" não detectadas. Métodos: amostra composta por 140 pacientes em acompanhamento em um ambulatório especializado no atendimento a pessoas com doenças crônicas não transmissíveis (DCNT) de um hospital universitário. As IMp foram identificadas e classificadas no DRUG-REAX® System e em oito sistemas de rastreamento de acesso aberto. As IMp de gravidade "Contraindicada" e "Maior" foram analisadas segundo o impacto clínico. Utilizou-se estatística descritiva e calculou-se sensibilidade e especificidade dos sistemas de rastreamento na identificação das IMp. Resultados: Os sistemas de acesso aberto pertencentes as bases Drugs.com, UCLA School of Health e CVC Caremark apresentaram sensibilidade e especificidade > 70%. A totalidade dos sistemas de acesso aberto não detectou os pares ciprofibrato + estatinas e metformina + sitagliptina, cujos impactos clínicos incluíram risco de miopatia e rabdomiólise e hipoglicemia, respectivamente. Cerca de um terço (37,5%) dos sistemas de acesso aberto não detectou a IMp ácido acetilsalicílico + hidroclorotiazida, capaz de ocasionar nefrotoxicidade. Conclusão: A maioria dos pares de IMp integra o rol terapêutico de pacientes com DCNT e cujos impactos clínicos são tempo-dependentes. A combinação de julgamento clínico, revisão periódica do plano terapêutico e os atributos de precisão (sensibilidade e especificidade) são fundamentais para garantir a segurança do paciente, sobretudo no contexto ambulatorial. (AU)

This study aims to evaluate the sensitivity and specificity of open-access screening systems in detecting potential drug-drug interactions (PDDIs) compared to the DRUG-REAX® system and analyze the potential clinical impact of PDDIs of "Contraindicated" and "Major" severities not detected. A cross-sectional study was conducted in an outpatient clinic specialized in caring for patients with noncommunicable diseases (NCDs) of a university hospital. PDDIs were identified and classified in the DRUG-REAX® System and eight open-access screening systems. The "Contraindicated" and "Major" severity PDDIs were analyzed according to clinical impact. Descriptive statistics were used and the sensitivity and specificity of the screening systems were calculated to identify the PDDIs. Results: The open-access systems Drugs.com, UCLA School of Health and CVC Caremark showed sensitivity and specificity > 70%. All open access systems did not detect the pairs ciprofibrate + statins and metformin + sitagliptin, whose clinical impacts included the risk of myopathy/ rhabdomyolysis and hypoglycemia, respectively. About a third (37.5%) of open-access systems did not detect PDDI acetylsalicylic acid + hydrochlorothiazide, which is capable of causing nephrotoxicity. Conclusion: Most pairs of PDDIs are part of the therapeutic role of patients with NCDs and whose clinical impacts are time-dependent. The combination of clinical judgment, periodic review of the therapeutic plan and the attributes of precision (sensitivity and specificity) are essential to ensure patient safety, especially in the outpatient setting. (AU)

Quality of the documentation of the Nursing process in clinical decision support systems / Qualidade da documentação do processo de enfermagem em sistemas de apoio à decisão clínica / Calidad de la documentación del proceso de Enfermería en sistemas de apoyo a la decisión clínica

Objective: to compare the quality of the Nursing process documentation in two versions of a clinical decision support system. Method: a quantitative and quasi-experimental study of the before-and-after type. The instrument used to measure the quality of the records was the Brazilian version of the Quality of Diagnoses, Interventions and Outcomes, which has four domains and a maximum score of 58 points. A total of 81 records were evaluated in version I (pre-intervention), as well as 58 records in version II (post-intervention), and the scores obtained in the two applications were compared. The interventions consisted of planning, pilot implementation of version II of the system, training and monitoring of users. The data were analyzed in the R software, using descriptive and inferential statistics. Results: the mean obtained at the pre-intervention moment was 38.24 and, after the intervention, 46.35 points. There was evidence of statistical difference between the means of the pre- and post-intervention groups, since the p-value was below 0.001 in the four domains evaluated. Conclusion: the quality of the documentation of the Nursing process in version II of the system was superior to version I. The efficacy of the system and the effectiveness of the interventions were verified. This study can contribute to the quality of documentation, care management, visibility of nursing actions and patient safety.

Objetivo: comparar a qualidade da documentação do processo de enfermagem em duas versões de um sistema de apoio à decisão clínica. Método: estudo quantitativo, quase-experimental do tipo antes e depois. O instrumento utilizado para mensurar a qualidade dos registros foi o Quality of Diagnoses, Interventions and Outcomes Versão Brasileira, que possui quatro domínios e escore máximo de 58 pontos. Foram avaliados 81 registros na versão I (pré-intervenção), 58 registros da versão II (pós-intervenção) e comparados os escores obtidos nas duas aplicações. As intervenções consistiram em planejamento, implantação piloto da versão II do sistema, treinamento e acompanhamento dos usuários. Os dados foram analisados no software R, utilizando-se estatística descritiva e inferencial. Resultados: a média obtida na pré-intervenção foi de 38,24 e pós-intervenção, 46,35 pontos. Houve evidências de diferença estatística entre as médias dos grupos pré e pós-intervenção, visto que o valor-p foi menor que 0,001 nos quatro domínios avaliados. Conclusão: a qualidade da documentação do processo de enfermagem na versão II do sistema foi superior à versão I. A eficácia do sistema e a efetividade das intervenções foram comprovadas. Este estudo pode contribuir para a qualidade da documentação, gerenciamento do cuidado, visibilidade das ações de enfermagem e segurança do paciente.

Objetivo: comparar la calidad de la documentación del proceso de Enfermería en dos versiones de un sistema de apoyo a la decisión clínica. Método: estudio cuantitativo y cuasi-experimental del tipo antes y después. El instrumento que se usó para medir la calidad de los registros fue la versión brasileña del Quality of Diagnoses, Interventions and Outcomes, que contiene cuatro dominios y puntaje máximo de 58 puntos. Se evaluaron 81 registros en la versión I (pre-intervención), 58 registros de la versión II (post-intervención) y se compararon las puntuaciones logradas en las dos aplicaciones. Las intervenciones consistieron en planificar, implementar de forma piloto la versión II del sistema, entrenar y realizar un seguimiento a los usuarios. Los datos se analizaron en el software R, utilizando estadística descriptiva e inferencial. Resultados: el promedio obtenido en el momento pre-intervención fue de 38,24 y en la post-intervención fue de 46,35 puntos. Se registraron evidencias de diferencia estadística entre los promedios de los grupos pre- y post-intervención, visto que el valor-p fue menor que 0,001 en los cuatro dominios evaluados. Conclusión: la calidad de la documentación del proceso de Enfermería en la versión II del sistema fue superior a la versión I. Se comprobó la eficacia del sistema y la efectividad de las intervenciones. Este estudio pudo contribuir para la calidad de la documentación, gerenciamiento de la atención, visibilidad de las acciones de Enfermería y seguridad del paciente.

Research progress in application of artificial intelligence in trauma assessment / 中华创伤杂志

Modern trauma scoring system tries hard to reflect and evaluate various injury severity and outcome in an objective，accurate and quick manner via digitalization so as to make correct diagnosis and treatment plan. However，current trauma scoring is mainly responsible for predicting death outcome，and it cannot reflect the changes of trauma rehabilitation function outcome in a dynamic way，thus making it difficult to meet clinical demands. In recent years，artificial intelligence technology and big data mining technology are helpful to integrate and analyze a large number of clinical data generated in the process of trauma treatment，and have a great application value for the continuous assessment of the whole trauma process. Such being the case，the authors sum up the research progress of physiology scoring，anatomy scoring，composite scoring of physiology and anatomy，functional scoring and artificial intelligence in trauma evaluation，attempting to provide references for building a new trauma scoring system.

Análise dos benefícios do sistema de apoio à decisão clínica em hospital para reduzir eventos adversos em casos de úlceras por pressão e quedas / Analysis of the benefit of the clinical decision support system in hospital to reduce adverse events in cases of pressure ulcer and falls / Análisis de los beneficios del sistema de apoyo a la decisión clínica en el hospital para reducir los eventos adversos en casos de úlceras por presión y caídas

Objetivo: O Objetivo desse estudo é analisar se sistemas de apoio à decisão integrados a prontuários eletrônicos apresentam benefícios para redução de eventos adversos em casos de úlceras por pressão e quedas em uma unidade de internação de um hospital. Métodos: É um estudo experimental tendo como base uma população de pacientes hospitalares internados e a utilização de métodos estatísticos. Os dados foram coletados do Sistema de Gestão de Eventos Adversos e do Sistema de Registro Eletrônico de Saúde em um período de 24 meses com uma amostra total de 106.658 registros de pacientes. Resultados: O uso de sistema de apoio à decisão nesse estudo trouxe contribuições para reduzir os casos de eventos adversos relacionados a úlceras por pressão, mas não apresentou contribuições relevantes em relação aos eventos adversos relacionados a quedas. Conclusão: Sistemas de apoio à decisão podem trazer contribuições importantes para a prestação de serviços assistenciais em pacientes internados.

Objective: This study aimed to analyze whether decision support systems integrated with electronic health record have benefits for reducing adverse events in cases of pressure ulcers and falls in an inpatient unit of a hospital. Methods: This is an experimental study based on an inpatient population and the use of statistical methods. Data were collected from both Adverse Event Management System and Electronic Health Records System from a 24-month period with a total sample of 106,658 patient records. Results: The use of a decision support system in this study brought contributions to reduce the cases of pressure ulcer related adverse events but did not present relevant contributions to reduce cases of fall related adverse events. Conclusion: Integrated decision support systems can add important contribution to healthcare services provided inside an inpatient unit.

Objetivo: El objetivo de este estudio es analizar si los sistemas de apoyo a la decisión integrados con registros médicos electrónicos tienen beneficios para reducir los eventos adversos en casos de úlceras por presión y caídas en una unidad de hospitalización de un hospital. Métodos: Este es un estudio experimental basado en una población de pacientes hospitalizados y el uso de métodos estadísticos. Los datos se recopilaron de lo Sistema de Gestión de Eventos Adversos y de lo Sistema de Registros de Salud Electrónicos un periodo de 24 meses con una muestra total de 106.658 registros de pacientes. Resultados: El uso del sistema de apoyo a la decisión evaluado ha contribuido para reducir los casos de eventos adversos relacionados con las úlceras por presión, pero no presentó contribuciones relevantes en relación con los eventos adversos relacionados con caídas. Conclusion: Los sistemas de apoyo a la decisión pueden hacer importantes contribuciones a la prestación de servicios de atención hospitalaria.

Current applications of artificial intelligence for intraoperative decision support in surgery / 医学前沿

Research into medical artificial intelligence (AI) has made significant advances in recent years, including surgical applications. This scoping review investigated AI-based decision support systems targeted at the intraoperative phase of surgery and found a wide range of technological approaches applied across several surgical specialties. Within the twenty-one (n = 21) included papers, three main categories of motivations were identified for developing such technologies: (1) augmenting the information available to surgeons, (2) accelerating intraoperative pathology, and (3) recommending surgical steps. While many of the proposals hold promise for improving patient outcomes, important methodological shortcomings were observed in most of the reviewed papers that made it difficult to assess the clinical significance of the reported performance statistics. Despite limitations, the current state of this field suggests that a number of opportunities exist for future researchers and clinicians to work on AI for surgical decision support with exciting implications for improving surgical care.

Registros de deterioração clínica antes e depois da implantação de um sistema informatizado de alerta / Records of clinical deterioration before and after the implantation of a computerized warning system

Resumo: Trata-se de pesquisa integrante do projeto de pesquisa Inteligência Artificial e o Gerenciamento de pacientes com sepse: sua influência na tomada de decisão, intervenção e resultados de enfermagem, da linha de pesquisa Gerenciamento de Serviços de Saúde e Enfermagem. Teve como objetivo comparar os registros relacionados ao preenchimento do protocolo institucional de sepse de pacientes sem e com o apoio de um sistema informatizado de alertas. Pesquisa quantitativa, exploratória e comparativa com dados provenientes do prontuário eletrônico de um hospital público localizado na região sul do Brasil. Foram incluídos os registros da equipe multidisciplinar de cinco setores do hospital, para os quais foram abertos os protocolos no período de janeiro a maio de 2018 sem o sistema de alertas e no período de julho a novembro com a implantação do sistema de alertas. A amostra foi composta por 232 protocolos, sendo que destes, 23 foram abertos duas vezes, totalizando uma amostra final de 209 pacientes, 107 protocolos referentes ao período pré-implantação do sistema de alerta e 102 protocolos referentes ao período de pós-implantação do sistema de alertas. Foi elaborado um diagrama de classes e um dicionário de dados para otimizar a compreensão das variáveis utilizadas no procedimento de abertura do protocolo de sepse da instituição que posteriormente foram tabuladas em planilhas eletrônicas. Para as análises comparativas foi utilizado o Software R. À análise exploratória, o perfil demográfico é semelhante entre os grupos, com prevalência de pacientes do sexo masculino e idade média de 55 anos. Na análise inferencial, houve diferença significativa no período pré-implantação do sistema de alerta quanto a distribuição dos registros dos parâmetros qSOFA utilizados na abertura do protocolo de sepse, com a prevalência do nível de consciência e pressão arterial sistólica respectivamente (p=0,0001 e p=0,0004). No período pós-implantação do sistema de alerta foram identificados resultados significativos quanto à diminuição do número de óbitos (p=0,0134), aumento no tempo de internamento (p=0,0001) e aumento da média de verificação de sinais vitais por paciente (p<0,001). A implantação de um sistema de alertas para identificação precoce de pacientes em deterioração clínica depende de fatores que não se limitam apenas à tecnologia. Engajar os profissionais quanto ao seu comprometimento nos registros eletrônicos e em reconhecer o potencial do uso de sistemas informatizados para a qualidade do cuidado hospitalar são aspectos que podem contribuir para o sucesso do uso dessas ferramentas de tecnologia na área da saúde.

Abstract: This research is part of the research project Artificial Intelligence and the Management of Patients with Sepsis: influence on nursing decision making, intervention, and outcomes, from the Health Services and Nursing Management research line. The aim was to compare the records related to the filling of the institutional sepsis protocol of patients with and without the support of a computerized alert system. Quantitative, exploratory and comparative research with data from the electronic medical records of a public hospital located in southern Brazil. The records of the multidisciplinary team from five ward of the hospital were included, which protocols were opened from January to May 2018 without the alert system and from July to November with the implementation of the alert system. The sample consisted of 232 protocols, which 23 were opened twice, reaching a final sample of 209 patients, 107 protocols referring to the pre-implantation of the alert system and 102 protocols referring to the postimplantation system. alerts. A class diagram and data dictionary were drawn up to optimize the understanding of the variables used in the institution's sepsis protocol opening procedure and later tabulated in spreadsheets. For comparative analysis, the R software was used. Exploratory analysis showed that the demographic profile is similar between the groups, male patients were prevalent in the average age of 55 years. Inferential analysis showed a significant difference in the pre-implantation of the alert system regarding the distribution of the records of the qSOFA parameters used at the opening of the sepsis protocol, with prevalence of consciousness level and systolic blood pressure respectively (p = 0.0001 and p = 0.0004). In the postimplantation of the alert system, significant results were identified regarding the decrease in the number of deaths (p = 0.0134), the increase in the length of stay (p = 0.0001) and the increase in the mean of vital signs verification per patient (p = 0.000000000000115). The implementation of an alert system for early identification of patients with clinical deterioration depends on factors that are not limited to technology alone. Engaging professionals in their commitment to electronic records and in recognizing the potential of using computerized systems for the quality of hospital care can contribute to the successful use of these technology tools in healthcare.

Development and enlightenment of intelligentized diagnostic and therapeutic mode for hepatobiliary tumors / 临床肝胆病杂志

Clinical decision support system (CDSS) and multidisciplinary team (MDT) are important measures to improve the efficiency and quality of healthcare service. This article reviews and analyzes the current status of the development of CDSS and MDT in China and foreign countries and summarizes their features and shortcomings, and with the background of the new era of “medical big data resources” and “big data technology”, this article looks into the future of developing an intelligentized diagnostic and therapeutic mode (a combination of CDSS and MDT) for hepatobiliary tumors suitable for the national conditions of China.