Factors causing Covid-19 disease and hazard function using multiple period logit model Available to Purchase

Corona virus disease-2019 abbreviated by Covid-19 is an infectious illness that initially occurred in Wuhan, China, in December 2019. This disease is caused by Corona virus that causes sick in animal or human. Corona virus causes the disease Covid-19. Because it has caused many sufferers around the world, this illness is considered a pandemic. The respiratory tract is an area that is easily affected by Covid-19. This infectious disease can spread when the sufferer cough, sneeze, and talk. Some of wall of respiratory tract will be released along with a splash of saliva called droplet. This research wants to find out the causes of the Covid-19 disease. Statistics related to this problem is survival analysis. The data analyzed is incidence of Covid-19 patient who are hospitalized. Because there are events with different starting times, type 3 data censorship is used. Data changes are logged every day. Data taken is data with uncensored status. This data is data when the patient has left the hospital with a recovered condition. The function related to time of hospitalization is the survival function, while the function related to censored and uncensored time is the hazard function. Regression model using simple period logit and multiple period logit. Simple period logit is logit function of the univariate hazard function which is a simple linear regression function with a time period. Meanwhile, multiple period logit is a development of the simple period logit. It is a logit function of the multivariate hazard function which is a multiple linear regression function with time periods. This hazard function is useful for determining the odds ratio. Research variables consist of response variable, indicator and covariates. The response variable is survival time, and the indicator variable is patient status, while the covariates are age, gender, symptom, systolic, diastolic, pulse, respiration, temperature, saturation, comorbid, and smoker. the Kaplan-Meier survival curve indicated that the longer patient is treated with treatment, the greater chance of patient being declared cured. Younger patients have a relatively faster recovery than older patients. Patients with asymptomatic conditions have a higher chance of recovering from the disease compared to symptomatic patients. Simple period logit modeling gives result that significant covariates on healing factor of Covid-19 patient are age and symptom covariates. But in multiple period logit modeling resulted in the factors causing the healing of covid-19 patients are age, symptom, respiration, and comorbid. The best hazard function is affected by covariates of age, symptom, respiration, and comorbid. Therefore, factors associated with Covid-19 are age, symptom, respiration, and comorbid. Factors of young age, asymptomatic, normal respiration, and no comorbidities have an effect on chance of recovering faster in patients with covid-19.