Navegando por Palavras-chave "Acurácia dos dados"

Agora exibindo 1 - 2 de 2
  • Imagem de Miniatura
    Item
    Acesso aberto (Open Access)
    Acurácia da Ecografia Vascular com Doppler comparado à Angiotomografia para o diagnóstico de estenose de carótida interna extracraniana
    (Universidade Federal de São Paulo, 2023-08-02) Daolio, Raul Muffato [UNIFESP]; Flumignan, Ronald Luiz Gomes [UNIFESP]; Nakano, Luis Carlos Uta [UNIFESP]; http://lattes.cnpq.br/7498955151131694; http://lattes.cnpq.br/6751410694690699; http://lattes.cnpq.br/1354091439126909
    Introdução: Estenoses de artéria carótida interna (ACI) são causas importantes de acidentes vasculares cerebrais (AVCs). A Ecografia Vascular com Doppler (EVD) costuma ser o primeiro exame a ser realizado durante sua investigação, mas possui acurácia variável na literatura e é comum complementar o estudo com a realização de outro exame quando frente a uma estenose com possibilidade de tratamento cirúrgico. Há carência de estudos de acurácia comparando a EVD com angiotomografia (ATC) na literatura. Métodos: Foi realizado um estudo de acurácia no qual se comparou a EVD com as ATCs de carótida de pacientes da EPM/UNIFESP realizadas em um mesmo período. Incluiu-se pacientes que realizaram ambos exames em um período de até 3 meses. As imagens foram avaliadas independentemente por 2 cirurgiões vasculares e, se houvesse divergência entre eles, um terceiro arbitrou o resultado. Realizou-se análise estatística determinando a acurácia, sensibilidade e especificidade de estenoses nas faixas de 50-99% e 70-99%. Também foi realizada a análise de discordância entre avaliadores de um mesmo exame para se comparar uma técnica à outra. Resultados: Foram incluídos nas análises estatísticas 45 pacientes e 84 artérias. Para estenoses 50-99% obteve-se acurácia de 69%, sensibilidade 89% e especificidade 63%. Para estenoses 70-99% obteve-se acurácia de 84%, sensibilidade de 61% e especificidade de 93%. A área sob a curva ROC é de 0,868. Ocorreu discordância entre avaliadores de ATC com mudança de conduta clínica para cirúrgica em pelo menos 37,5% dos laudos conflitantes. Conclusão: A acurácia da EVD para estenoses entre 50-99% é de 69%. Para estenoses entre 70-99% a acurácia é de 84%. A área sob a curva ROC é de 0,868. A EVD pode ser usada exclusivamente para indicar endarterectomia de carótida desde que seja realizado um segundo exame para confirmação. Análises de ATC podem ser dependentes do avaliador a ponto de mudar a conduta.
  • Imagem de Miniatura
    Item
    Acesso aberto (Open Access)
    Análise de sobrevida e fatores prognósticos de idosas com câncer de mama em um hospital público de referência em São Paulo
    (Universidade Federal de São Paulo (UNIFESP), 2018-12-20) Souza, Camila Brandao de [UNIFESP]; Schirmer, Janine [UNIFESP]; Fustinoni, Suzete Maria [UNIFESP]; http://lattes.cnpq.br/2016866886536539; São Paulo; http://lattes.cnpq.br/3721636964139813; http://lattes.cnpq.br/8426650565647334; Universidade Federal de São Paulo (UNIFESP)
    Objectives: To assess the completeness and the tendency to completeness of the records data; Survival rate analyze, and the prognostic factors of elderly women with breast cancer, diagnosed and assisted in the period from 2001 to 2006 in a public hospital, a reference centre for women’s health in the state of Sao Paulo. Method: Two studies were performed. The first was an analytic study based on secondary data, with a population composed of all elderly women diagnosed in the service in the period of the study, in a total of 1241 registries. For the noncompleteness analysis, the following classification was used: excellent (<5%), good (510%), regular (1020%), bad (2050%) and very bad (≥50%). The second study was a retrospective cohort of secondary data collected from the Hospital’s Cancer Registry and its records. The KaplanMeier method was used for an estimate of the survival rate and the logrank to test the statistical significance, comparing the survival curves. To verify the independent effect of the statistical significant variables of up to 10%, the Cox multivariate proportional risk method was used (Hazards Ratio). Results: Regarding the completeness of the data, the socioeconomical and demographic variables, as well as the risk and behavioral factors were the predominant ones among the ones classified as regular, bad and very bad. The best scores were the ones for posttreatment variables, followed by the ones related to diagnosis and treatment. The only variable with a decreasing tendency of noncompleteness was that of breast cancer family history (p = 0,05). The variables with an increasing noncompleteness tendency were: race/color (p = 0,01), years of education (p = 0,01), oral contraceptives use (p = 0,002), oral contraceptives time of use (p = 0,002), hormonal replenishment (p = 0,007) and breast feeding (p = 0,004). The global survival rates at the end of the analyzed period were that of 56%, and by specific cause, 72%. The variables of age 8084 years old (p = 0,001), late stage diagnosis (p = 0,001), ≥ 4cm tumors (p = 0,001), undermined surgical margins (p = 0,001), progesterone negative tumors (p = 0,028), p53 positive tumors (p = 0,034), grade III nuclear (p = 0,011) and histological grade 3 (p = 0,007), the presence of tumor necrosis (p = 0,001), 4 or more lymph nodes undermined lymph nodes (p = 0,001), the occurrence of locoregional relapse (p = 0,001) or presence of metastasis (p = 0,001) were significantly related to the number of lost years of life and to the accumulated survival rate. In the model, after adjusted hazard ratio, the following were maintained: age group (HR: 5,96; IC95%: 2,5713,85), size of tumor (HR: 3,42; IC95%: 1,179,99), number of undermined lymph nodes (HR: 2,00; IC95%: 1,043,87), and metastasis (HR: 3,31; IC95%: 1,895,82). Conclusion: Among the variables that were classified as regular, bad and very bad, the tendency to completeness predominated as constant, with 12 registries, followed by the growing tendency of noncompleteness, with 6 registries, and only the variable family history of breast cancer has presented an improvement in the tendency of completeness over the years. The complete registration of data in the forms is an inherent task of the whole of the healthcare team, crucial to establish the assistance protocols, for the development of research, as well as in the implementation of public health policies. The prognosis and survival rate fall as age, the size of the tumor, the number rise of compromised lymph nodes, and metastasis happens. The diagnosis in early stages for women is a crucial need.