A carregar...
Publicação
Estudo do impacto da pandemia nas estruturas residenciais para idosos de Portugal em 2020
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 521.42 KB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
O nosso projeto baseia-se na análise dos registos de casos e mortes por COVID-19 em utentes
ocorridos em ERPI entre 15 de abril e 27 de dezembro de 2020. Com base nestes dados,
determinou-se Risco Relativo (RR) de infeção e de morte dos utentes em ERPIs relativamente
à população idosa em geral para as cinco regiões de saúde (com a exceção da região Alentejo
devido a problemas de qualidade nos seus registos de casos e óbitos), e para os dois períodos
de 2020 que correspondem à primeira e à segunda vaga daquele ano. Verificámos que em todas
as regiões o RR de infeção foi reduzido na segunda vaga, o que demonstra a capacidade de
adaptação das ERPIs em geral no sentido de melhorar a proteção dos utentes. Esta análise
mostra também que em Lisboa e Vale do Tejo (LVT) surge como um outlier de excesso de
RR e de menor redução de RR na segunda vaga. Isto motivou um estudo mais detalhado em
que LVT foi comparada com o conjunto das outras regiões (OUTRAS). Fizemos então uma
análise por tamanho das ERPIs, depois de verificar que a distribuição das ERPIs por tamanhos
é aproximadamente a mesma em LVT e em OUTRAS. Os indicadores analisados em função do
tamanho foram o odds ratio (OR) de um lar registar um ou mais casos entre os seus utentes
e o tamanho de cada surto registado (TS) medido como a percentagem dos utentes infetados.
A métrica TS comporta-se em todas as regiões de acordo com o esperado e o reportado na
literatura, diminuindo com o tamanho da ERPI. Quanto ao OR, em OUTRAS ele comporta-se
como esperado, aumentando significativamente com o tamanho da ERPI, mas em LVT o OR
mínimo ocorre nas ERPIs médias, e as ERPIs pequenas surgem com um nível de risco muito
elevado. Assim, o comportamento de outlier de LVT resulta em parte de um excesso de risco
de infeção nas ERPIs pequenas. A análise do ratio funcionário utente mostra também que as
ERPIs pequenas de LVT são outliers com valores muito pequenos desse ratio, relativamente às
outras classes de ERPIs.
Our project is based on the analysis of records of COVID-19 cases and deaths among residents of nursing homes that occurred between April 15th and December 27th, 2020. Based on these data, the relative risk (RR) of infection and death of residents in nursing homes was determined relative to the general elderly population for the five health regions (excluding the Alentejo region due to problems with the quality of its case and death records), and for the two periods of 2020 corresponding to the first and second waves of that year. We found that in all regions the RR of infection was reduced in the second wave, which demonstrates the ability of nursing homes in general to adapt in order to better protect residents. This analysis also shows that Lisbon and Vale do Tejo (LVT) appears as an outlier with a high RR and a lower reduction in RR in the second wave. This motivated a more detailed study in which LVT was compared to the remaining regions (OUTRAS). We then carried out analysis by size of nursing homes after verifying that the distribution of nursing homes by size is approximately the same in LVT and in OUTRAS. The indicators analyzed based on size were the odds ratio (OR) of a nursing home having one or more cases among its residents and the size of each outbreak recorded (TS) measured as the percentage of infected residents. The TS metric behaves in all regions as expected and reported in the literature, decreasing with the size of the nursing home. As for the OR, in OUTRAS it behaves as expected, significantly increasing with the size of the nursing home, but in LVT the minimum OR occurs in medium-sized nursing homes, and small nursing homes have a very high risk level. Thus, LVT’s outlier behavior is partially due to an excess risk of infection in small nursing homes. The analysis of the staff-to-resident ratio also shows that small nursing homes in LVT are outliers with very low values of this ratio relative to other classes of nursing homes.
Our project is based on the analysis of records of COVID-19 cases and deaths among residents of nursing homes that occurred between April 15th and December 27th, 2020. Based on these data, the relative risk (RR) of infection and death of residents in nursing homes was determined relative to the general elderly population for the five health regions (excluding the Alentejo region due to problems with the quality of its case and death records), and for the two periods of 2020 corresponding to the first and second waves of that year. We found that in all regions the RR of infection was reduced in the second wave, which demonstrates the ability of nursing homes in general to adapt in order to better protect residents. This analysis also shows that Lisbon and Vale do Tejo (LVT) appears as an outlier with a high RR and a lower reduction in RR in the second wave. This motivated a more detailed study in which LVT was compared to the remaining regions (OUTRAS). We then carried out analysis by size of nursing homes after verifying that the distribution of nursing homes by size is approximately the same in LVT and in OUTRAS. The indicators analyzed based on size were the odds ratio (OR) of a nursing home having one or more cases among its residents and the size of each outbreak recorded (TS) measured as the percentage of infected residents. The TS metric behaves in all regions as expected and reported in the literature, decreasing with the size of the nursing home. As for the OR, in OUTRAS it behaves as expected, significantly increasing with the size of the nursing home, but in LVT the minimum OR occurs in medium-sized nursing homes, and small nursing homes have a very high risk level. Thus, LVT’s outlier behavior is partially due to an excess risk of infection in small nursing homes. The analysis of the staff-to-resident ratio also shows that small nursing homes in LVT are outliers with very low values of this ratio relative to other classes of nursing homes.
Descrição
Tese de Mestrado, Bioinformática e Biologia Computacional, 2023, Universidade de Lisboa, Faculdade de Ciências
Palavras-chave
Covid-19 Risco Relativo Estruturas Residenciais de Pessoas Idosas Estudo Observacional de Coorte Aprendizados durante a pandemia Teses de mestrado - 2023