| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| DM_MiguelMaia_MMADE_2022 | 1.2 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
Atualmente, um dos principais problemas encarados pelas instituições de saúde, a
nível mundial, é o absentismo dos pacientes perante as consultas marcadas. O número
de consultas e/ou exames marcados e não realizados, chegam a atingir níveis elevados,
sendo este fenómeno visível em vários países e em diferentes tipos de especialidades e
atendimentos. Este comportamento gera às instituições: perda de receita, desperdícios
de recursos, desorganização da oferta de serviços e limitação dos utentes na garantia
do atendimento para a sua assistência. Adicionalmente, gera o uso inadequado da
oferta, intensifica os tempos de espera e as respetivas filas, além de acarretar um
prejuízo financeiro. Por isso, torna-se importante prever o não comparecimento dos
pacientes, assim como perceber o perfil destes pacientes ausentes para que seja
possível criar mecanismos que o mitiguem e descobrir as razões que os levam a não
comparecer às consultas marcadas. Adicionalmente, a pandemia originada pela COVID-
19 apresentou-se como um desafio à gestão hospitalar que rapidamente se teve de
adaptar à nova realidade. Como tal, este estudo apresenta dois objetivos principais: (i)
entender os fatores associados ao risco de absentismo de pacientes e (ii) entender se
houve alterações nos perfis os pacientes perante o risco de absentismo após o início da
pandemia.
Assim, este trabalho apresenta a aplicação do modelo logístico binário, identificado
na literatura como um dos modelos mais utilizados e eficazes, para ajudar a prever se o
paciente comparecerá ou não à consulta agendada. Permitindo entender quais os
fatores relacionados com o risco de faltar e comparando os perfis dos pacientes em dois
horizontes temporais distintos: pré-pandemia e durante a pandemia, e expondo
sugestões de melhoria para a diminuição das taxas de absentismo.
Com base em estudos anteriores, foram incluídos todos os fatores de risco
significativos ao absentismo de utentes, tais como: idade, género, estado civil, distância
do hospital até à residência do paciente, número de consultas anteriores, primeira
consulta/acompanhamento, mês, dia da semana da consulta, precipitação, velocidade
do vento, tempo de espera, intervalo entre consultas, número de não comparências
anteriores e estação do ano; Por último, o modelo de regressão logístico permitiu identificar diferentes perfis de
pacientes por diferentes especialidades, assim o hospital deverá considerar um modelo
logístico distinto por cada tipo de especialidade. O Covid-19 provocou mudanças no
comportamento dos pacientes resultando num absentismo destes às consultas. Os
fatores de risco comuns identificados nos modelos antes e durante a pandemia, foram
o tempo de espera, o número de não comparências anteriores e o número de consultas
anteriores.
Relativamente aos fatores de risco característicos do período anterior à pandemia,
destacam-se o estado civil e a idade, enquanto durante a pandemia realça-se o mês da
consulta.
Nowadays, one of the main problems faced by health institutions worldwide is healthcare patients not attending to their scheduled appointments. The number of appointments and/or exams scheduled and not performed can reach high levels, and this phenomenon is visible in several countries and in different types of specialties and treatments. This behavior generates loss of revenue, waste of resources, disorganizes the services, and limits the assurance of healthcare to the patients. All this negative behavior generates inadequate use of services, intensifies waiting times and queues, and causes financial loss. Therefore, it is important to predict the non-attendance of patients, as well as to understand the profile of these absent patients so that it is possible to create mechanisms to mitigate it and find out the reasons that lead them not to attend the scheduled appointments in the first place. Additionally, the pandemic originated by the COVID-19 virus also presented a challenge to hospital management, that quickly had to adapt to a new reality. Therefore, this study has two main goals: (i) to understand the factors associated with the risk of patient non-attendance and (ii) to understand if there were changes in the patients' profiles regarding the risk of nonattendance after the beginning of the pandemic. For this, this study will present the application of the binary logistic model - identified in the literature as one of the most effective and used model- to help predict whether or not the patient will attend the scheduled appointment. This will allow us to understand which factors are related to the risk of non-attendance and compare patient profiles in two distinct time horizons: pre-pandemic and during the pandemic, as well as setting out suggestions for improving the decrease of non-attendance rates. Based on previous studies, all significant risk factors for patient absenteeism were included, such as: age, gender, marital status, distance from the hospital to the patient's residence, number of previous visits, first visit/attendance, month, day of the week of the visit, rainfall, wind speed, waiting time, interval between visits, number of previous no-shows and season of the year. Finally, the logistic regression model allowed us to identify different patient profiles by different specialties, so the hospital should consider a distinct logistic model by each type of specialty. Covid-19 caused changes in patient behavior resulting in patient absenteeism from appointments. The common risk factors identified in the models before and during the pandemic were waiting time, number of prior no-shows, and number of prior visits. Regarding the risk factors characteristic of the period before the pandemic, marital status and age stand out, while during the pandemic the month of the visit stands out.
Nowadays, one of the main problems faced by health institutions worldwide is healthcare patients not attending to their scheduled appointments. The number of appointments and/or exams scheduled and not performed can reach high levels, and this phenomenon is visible in several countries and in different types of specialties and treatments. This behavior generates loss of revenue, waste of resources, disorganizes the services, and limits the assurance of healthcare to the patients. All this negative behavior generates inadequate use of services, intensifies waiting times and queues, and causes financial loss. Therefore, it is important to predict the non-attendance of patients, as well as to understand the profile of these absent patients so that it is possible to create mechanisms to mitigate it and find out the reasons that lead them not to attend the scheduled appointments in the first place. Additionally, the pandemic originated by the COVID-19 virus also presented a challenge to hospital management, that quickly had to adapt to a new reality. Therefore, this study has two main goals: (i) to understand the factors associated with the risk of patient non-attendance and (ii) to understand if there were changes in the patients' profiles regarding the risk of nonattendance after the beginning of the pandemic. For this, this study will present the application of the binary logistic model - identified in the literature as one of the most effective and used model- to help predict whether or not the patient will attend the scheduled appointment. This will allow us to understand which factors are related to the risk of non-attendance and compare patient profiles in two distinct time horizons: pre-pandemic and during the pandemic, as well as setting out suggestions for improving the decrease of non-attendance rates. Based on previous studies, all significant risk factors for patient absenteeism were included, such as: age, gender, marital status, distance from the hospital to the patient's residence, number of previous visits, first visit/attendance, month, day of the week of the visit, rainfall, wind speed, waiting time, interval between visits, number of previous no-shows and season of the year. Finally, the logistic regression model allowed us to identify different patient profiles by different specialties, so the hospital should consider a distinct logistic model by each type of specialty. Covid-19 caused changes in patient behavior resulting in patient absenteeism from appointments. The common risk factors identified in the models before and during the pandemic were waiting time, number of prior no-shows, and number of prior visits. Regarding the risk factors characteristic of the period before the pandemic, marital status and age stand out, while during the pandemic the month of the visit stands out.
Description
Keywords
absentismo regressão logística profiling agendamentos médicos horizontes temporais