Projeto covid1 regressão
Thomas Daher
30/01/2023
Instalação de pacotes e extração de dados
Criando tabela casos
Filtrando somente casos de covid
casos_fil_covid <- filter(casos_fil, CLASSI_FIN == 5)Criando a coluna menor data e a coluna epiweek(SEM_DTMIN)
Deixando nascimento em formato de data para o R e analisando a coluna NU_IDADE_N. Podemos ver que algumas pessoas possuem idades não condizentes com a realidade, com idades menores do que 0 ou muito acima da pessoa mais velha registrada no Brasil, portanto limito a 100 anos de idade.
casos_fil_covid$DT_NASC<- as_date(casos_fil_covid$DT_NASC, format= "%d/%m/%Y")
count(casos_fil_covid, NU_IDADE_N <= 0)## # A tibble: 2 x 2
## `NU_IDADE_N <= 0` n
## <lgl> <int>
## 1 FALSE 2134830
## 2 TRUE 1175count(casos_fil_covid, NU_IDADE_N >= 100)## # A tibble: 2 x 2
## `NU_IDADE_N >= 100` n
## <lgl> <int>
## 1 FALSE 2132815
## 2 TRUE 3190casos_fil_covid <- filter(casos_fil_covid, NU_IDADE_N > 0)
casos_fil_covid <- filter(casos_fil_covid, NU_IDADE_N < 100)Summary do df
summary(casos_fil_covid)## DT_NOTIFIC SEM_NOT SG_UF_NOT ID_MUNICIP
## Min. :2020-02-21 Min. : 1.00 Length:2131640 Length:2131640
## 1st Qu.:2020-11-11 1st Qu.:12.00 Class :character Class :character
## Median :2021-03-24 Median :21.00 Mode :character Mode :character
## Mean :2021-03-17 Mean :22.53
## 3rd Qu.:2021-06-15 3rd Qu.:30.00
## Max. :2022-12-09 Max. :53.00
##
## CO_MUN_NOT CS_SEXO DT_NASC CS_GESTANT
## Min. :110001 Length:2131640 Min. :1920-04-05 Min. :0.000
## 1st Qu.:310620 Class :character 1st Qu.:1948-09-13 1st Qu.:5.000
## Median :351880 Mode :character Median :1961-10-13 Median :6.000
## Mean :344123 Mean :1962-10-22 Mean :5.779
## 3rd Qu.:410690 3rd Qu.:1975-09-07 3rd Qu.:6.000
## Max. :530010 Max. :2022-11-22 Max. :9.000
## NA's :1839
## CS_RACA CS_ESCOL_N PAC_COCBO CS_ZONA
## Min. :1.00 Min. :0.0 Length:2131640 Min. :1.00
## 1st Qu.:1.00 1st Qu.:2.0 Class :character 1st Qu.:1.00
## Median :4.00 Median :4.0 Mode :character Median :1.00
## Mean :3.49 Mean :5.4 Mean :1.15
## 3rd Qu.:4.00 3rd Qu.:9.0 3rd Qu.:1.00
## Max. :9.00 Max. :9.0 Max. :9.00
## NA's :28517 NA's :715644 NA's :232617
## VACINA_COV VACINA DT_UT_DOSE HOSPITAL
## Min. :1.0 Min. :1.0 Length:2131640 Min. :1.00
## 1st Qu.:2.0 1st Qu.:2.0 Class :character 1st Qu.:1.00
## Median :2.0 Median :2.0 Mode :character Median :1.00
## Mean :2.6 Mean :5.1 Mean :1.04
## 3rd Qu.:2.0 3rd Qu.:9.0 3rd Qu.:1.00
## Max. :9.0 Max. :9.0 Max. :9.00
## NA's :325989 NA's :526559 NA's :45140
## DT_INTERNA UTI DT_ENTUTI
## Min. :2020-01-05 Min. :1.00 Min. :2020-01-05
## 1st Qu.:2020-11-06 1st Qu.:1.00 1st Qu.:2020-11-09
## Median :2021-03-21 Median :2.00 Median :2021-03-23
## Mean :2021-03-23 Mean :1.79 Mean :2021-03-17
## 3rd Qu.:2021-06-11 3rd Qu.:2.00 3rd Qu.:2021-06-14
## Max. :9202-09-11 Max. :9.00 Max. :4202-05-26
## NA's :253860
## DT_SAIDUTI CLASSI_FIN EVOLUCAO DT_EVOLUCA
## Min. :2020-02-21 Min. :5 Min. :1.0 Min. :2020-02-21
## 1st Qu.:2020-11-12 1st Qu.:5 1st Qu.:1.0 1st Qu.:2020-11-17
## Median :2021-03-26 Median :5 Median :1.0 Median :2021-03-31
## Mean :2021-03-18 Mean :5 Mean :1.5 Mean :2021-03-23
## 3rd Qu.:2021-06-16 3rd Qu.:5 3rd Qu.:2.0 3rd Qu.:2021-06-21
## Max. :2121-03-13 Max. :5 Max. :9.0 Max. :2022-12-04
## NA's :99271
## NU_IDADE_N NOSOCOMIAL AVE_SUINO FEBRE
## Min. : 1.00 Min. :1.0 Min. :1.0 Min. :1.0
## 1st Qu.:45.00 1st Qu.:2.0 1st Qu.:2.0 1st Qu.:1.0
## Median :59.00 Median :2.0 Median :2.0 Median :1.0
## Mean :57.91 Mean :2.6 Mean :3.3 Mean :1.4
## 3rd Qu.:72.00 3rd Qu.:2.0 3rd Qu.:2.0 3rd Qu.:2.0
## Max. :99.00 Max. :9.0 Max. :9.0 Max. :9.0
## NA's :374777 NA's :376972 NA's :340351
## TOSSE GARGANTA DISPNEIA DESC_RESP
## Min. :1.0 Min. :1 Min. :1.00 Min. :1.0
## 1st Qu.:1.0 1st Qu.:2 1st Qu.:1.00 1st Qu.:1.0
## Median :1.0 Median :2 Median :1.00 Median :1.0
## Mean :1.3 Mean :2 Mean :1.28 Mean :1.4
## 3rd Qu.:1.0 3rd Qu.:2 3rd Qu.:1.00 3rd Qu.:2.0
## Max. :9.0 Max. :9 Max. :9.00 Max. :9.0
## NA's :268697 NA's :615677 NA's :265435 NA's :409211
## SATURACAO DIARREIA VOMITO DOR_ABD
## Min. :1.0 Min. :1 Min. :1.0 Min. :1.0
## 1st Qu.:1.0 1st Qu.:2 1st Qu.:2.0 1st Qu.:2.0
## Median :1.0 Median :2 Median :2.0 Median :2.0
## Mean :1.4 Mean :2 Mean :2.1 Mean :2.2
## 3rd Qu.:2.0 3rd Qu.:2 3rd Qu.:2.0 3rd Qu.:2.0
## Max. :9.0 Max. :9 Max. :9.0 Max. :9.0
## NA's :341842 NA's :643986 NA's :671957 NA's :873571
## FADIGA PERD_OLFT PERD_PALA OUTRO_SIN
## Min. :1.0 Min. :1.0 Min. :1.0 Min. :1.0
## 1st Qu.:1.0 1st Qu.:2.0 1st Qu.:2.0 1st Qu.:1.0
## Median :2.0 Median :2.0 Median :2.0 Median :2.0
## Mean :1.9 Mean :2.1 Mean :2.1 Mean :1.8
## 3rd Qu.:2.0 3rd Qu.:2.0 3rd Qu.:2.0 3rd Qu.:2.0
## Max. :9.0 Max. :9.0 Max. :9.0 Max. :9.0
## NA's :796163 NA's :852465 NA's :853816 NA's :617519
## OUTRO_DES FATOR_RISC PUERPERA CARDIOPATI
## Length:2131640 Length:2131640 Min. :1.0 Min. :1.0
## Class :character Class :character 1st Qu.:2.0 1st Qu.:1.0
## Mode :character Mode :character Median :2.0 Median :1.0
## Mean :2.2 Mean :1.4
## 3rd Qu.:2.0 3rd Qu.:2.0
## Max. :9.0 Max. :9.0
## NA's :1352981 NA's :1099368
## HEMATOLOGI SIND_DOWN HEPATICA ASMA
## Min. :1.0 Min. :1.0 Min. :1.0 Min. :1.0
## 1st Qu.:2.0 1st Qu.:2.0 1st Qu.:2.0 1st Qu.:2.0
## Median :2.0 Median :2.0 Median :2.0 Median :2.0
## Mean :2.2 Mean :2.2 Mean :2.2 Mean :2.1
## 3rd Qu.:2.0 3rd Qu.:2.0 3rd Qu.:2.0 3rd Qu.:2.0
## Max. :9.0 Max. :9.0 Max. :9.0 Max. :9.0
## NA's :1350649 NA's :1353343 NA's :1352960 NA's :1340898
## DIABETES NEUROLOGIC PNEUMOPATI IMUNODEPRE
## Min. :1.0 Min. :1.0 Min. :1.0 Min. :1.0
## 1st Qu.:1.0 1st Qu.:2.0 1st Qu.:2.0 1st Qu.:2.0
## Median :2.0 Median :2.0 Median :2.0 Median :2.0
## Mean :1.6 Mean :2.1 Mean :2.1 Mean :2.1
## 3rd Qu.:2.0 3rd Qu.:2.0 3rd Qu.:2.0 3rd Qu.:2.0
## Max. :9.0 Max. :9.0 Max. :9.0 Max. :9.0
## NA's :1172542 NA's :1331091 NA's :1332442 NA's :1342756
## RENAL OBESIDADE OBES_IMC OUT_MORBI
## Min. :1.0 Min. :1 Length:2131640 Min. :1.0
## 1st Qu.:2.0 1st Qu.:2 Class :character 1st Qu.:1.0
## Median :2.0 Median :2 Mode :character Median :1.0
## Mean :2.1 Mean :2 Mean :1.5
## 3rd Qu.:2.0 3rd Qu.:2 3rd Qu.:2.0
## Max. :9.0 Max. :9 Max. :9.0
## NA's :1334452 NA's :1303592 NA's :1173521
## MORB_DESC DOSE_1_COV DOSE_2_COV DOSE_REF
## Length:2131640 Length:2131640 Length:2131640 Length:2131640
## Class :character Class :character Class :character Class :character
## Mode :character Mode :character Mode :character Mode :character
##
##
##
##
## FAB_COVREF ANTIVIRAL TP_ANTIVIR OUT_ANTIV
## Length:2131640 Min. :1.0 Min. :1.0 Length:2131640
## Class :character 1st Qu.:2.0 1st Qu.:1.0 Class :character
## Mode :character Median :2.0 Median :1.0 Mode :character
## Mean :3.2 Mean :1.2
## 3rd Qu.:2.0 3rd Qu.:1.0
## Max. :9.0 Max. :3.0
## NA's :347443 NA's :2008382
## DT_ANTIVIR SUPORT_VEN RAIOX_RES RAIOX_OUT
## Length:2131640 Min. :1.00 Min. :1.0 Length:2131640
## Class :character 1st Qu.:2.00 1st Qu.:3.0 Class :character
## Mode :character Median :2.00 Median :6.0 Mode :character
## Mean :2.26 Mean :5.3
## 3rd Qu.:2.00 3rd Qu.:6.0
## Max. :9.00 Max. :9.0
## NA's :256513 NA's :857548
## TOMO_RES TOMO_OUT AMOSTRA TP_AMOSTRA
## Min. :1.0 Length:2131640 Min. :1.00 Min. :1.00
## 1st Qu.:1.0 Class :character 1st Qu.:1.00 1st Qu.:1.00
## Median :1.0 Mode :character Median :1.00 Median :1.00
## Mean :3.1 Mean :1.08 Mean :1.38
## 3rd Qu.:6.0 3rd Qu.:1.00 3rd Qu.:1.00
## Max. :9.0 Max. :9.00 Max. :9.00
## NA's :862861 NA's :71794 NA's :184777
## OUT_AMOST PCR_RESUL DT_PCR POS_PCRFLU
## Length:2131640 Min. :1.00 Length:2131640 Min. :1.0
## Class :character 1st Qu.:1.00 Class :character 1st Qu.:2.0
## Mode :character Median :1.00 Mode :character Median :2.0
## Mean :1.89 Mean :2.8
## 3rd Qu.:2.00 3rd Qu.:2.0
## Max. :9.00 Max. :9.0
## NA's :195201 NA's :1317296
## TP_FLU_PCR PCR_FLUASU DT_MIN ANO_DTMIN
## Min. :1.0 Min. :1.0 Min. :2020-01-05 Min. :2020
## 1st Qu.:1.0 1st Qu.:2.0 1st Qu.:2020-11-05 1st Qu.:2020
## Median :1.0 Median :2.0 Median :2021-03-21 Median :2021
## Mean :1.1 Mean :2.5 Mean :2021-03-12 Mean :2021
## 3rd Qu.:1.0 3rd Qu.:3.0 3rd Qu.:2021-06-10 3rd Qu.:2021
## Max. :2.0 Max. :6.0 Max. :2022-12-04 Max. :2022
## NA's :2130656 NA's :2130801
## SEM_DTMIN
## Min. : 1.00
## 1st Qu.: 45.00
## Median : 65.00
## Mean : 63.42
## 3rd Qu.: 76.00
## Max. :154.00
## Criando tabelas por tipo (Possuimos muitas colunas, algumas delas com muitos nulos, portanto para verificar quais possuem impacto na mortalidade vou separar por tipo, por exemplo, demográficas, comorbidades, relacionadas a vacinação, etc.)
Demográficas e economicas
casos_fil_covid_dem <- casos_fil_covid %>% select(DT_MIN, SG_UF_NOT, ID_MUNICIP, CO_MUN_NOT, CS_SEXO, CS_GESTANT, CS_RACA, CS_ESCOL_N, PAC_COCBO, CS_ZONA, NU_IDADE_N, EVOLUCAO)
summary(casos_fil_covid_dem)## DT_MIN SG_UF_NOT ID_MUNICIP CO_MUN_NOT
## Min. :2020-01-05 Length:2131640 Length:2131640 Min. :110001
## 1st Qu.:2020-11-05 Class :character Class :character 1st Qu.:310620
## Median :2021-03-21 Mode :character Mode :character Median :351880
## Mean :2021-03-12 Mean :344123
## 3rd Qu.:2021-06-10 3rd Qu.:410690
## Max. :2022-12-04 Max. :530010
##
## CS_SEXO CS_GESTANT CS_RACA CS_ESCOL_N
## Length:2131640 Min. :0.000 Min. :1.00 Min. :0.0
## Class :character 1st Qu.:5.000 1st Qu.:1.00 1st Qu.:2.0
## Mode :character Median :6.000 Median :4.00 Median :4.0
## Mean :5.779 Mean :3.49 Mean :5.4
## 3rd Qu.:6.000 3rd Qu.:4.00 3rd Qu.:9.0
## Max. :9.000 Max. :9.00 Max. :9.0
## NA's :28517 NA's :715644
## PAC_COCBO CS_ZONA NU_IDADE_N EVOLUCAO
## Length:2131640 Min. :1.00 Min. : 1.00 Min. :1.0
## Class :character 1st Qu.:1.00 1st Qu.:45.00 1st Qu.:1.0
## Mode :character Median :1.00 Median :59.00 Median :1.0
## Mean :1.15 Mean :57.91 Mean :1.5
## 3rd Qu.:1.00 3rd Qu.:72.00 3rd Qu.:2.0
## Max. :9.00 Max. :99.00 Max. :9.0
## NA's :232617 NA's :99271Estruturando em one hot encoded para poder fazer a regressão logit.
casos_fil_covid_dem <- dummy_cols(casos_fil_covid_dem, select_columns = c('CS_SEXO', 'CS_GESTANT', 'CS_RACA', 'CS_ZONA'),
remove_selected_columns = TRUE)
casos_fil_covid_dem$EVOLUCAO <- if_else(casos_fil_covid_dem$EVOLUCAO == 1 , 1, 0)Porcentagem de nulos na coluna EVOLUCAO.
sum(is.na(casos_fil_covid$EVOLUCAO)) / length(casos_fil_covid$EVOLUCAO)## [1] 0.04657025Variáveis selecionadas: CS_SEX_F = se é mulher ou não, CS_RACA_2 = se é preto ou não, CS_ZONA_2 = se é rural ou não, NU_IDADE_N = idade da pessoa. EVOLUCAO = 1 para sobreviveu, 0 para óbito. Recordo também que o comando glm omite as linhas com valores nulos.
casos_fil_covid.logit <- glm(formula = EVOLUCAO ~ CS_SEXO_F + CS_RACA_2 + CS_ZONA_2 + NU_IDADE_N, family = binomial(link = "logit"),
data = casos_fil_covid_dem)
summary(casos_fil_covid.logit)##
## Call:
## glm(formula = EVOLUCAO ~ CS_SEXO_F + CS_RACA_2 + CS_ZONA_2 +
## NU_IDADE_N, family = binomial(link = "logit"), data = casos_fil_covid_dem)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.4794 -1.1620 0.6629 0.9259 1.7909
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 2.921e+00 6.452e-03 452.77 <2e-16 ***
## CS_SEXO_F 1.441e-01 3.330e-03 43.26 <2e-16 ***
## CS_RACA_2 -3.064e-01 7.815e-03 -39.20 <2e-16 ***
## CS_ZONA_2 -1.410e-01 7.443e-03 -18.95 <2e-16 ***
## NU_IDADE_N -3.892e-02 9.946e-05 -391.31 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 2322208 on 1795884 degrees of freedom
## Residual deviance: 2140628 on 1795880 degrees of freedom
## (335755 observations deleted due to missingness)
## AIC: 2140638
##
## Number of Fisher Scoring iterations: 3Variáveis relacionadas à vacinação
rm(casos_fil_covid_dem)
gc()## used (Mb) gc trigger (Mb) max used (Mb)
## Ncells 5652072 301.9 12354841 659.9 5714417 305.2
## Vcells 540287406 4122.1 948513124 7236.6 948401121 7235.8casos_fil_covid_vacina <- casos_fil_covid %>% select(DT_MIN, VACINA_COV, VACINA, NU_IDADE_N, EVOLUCAO)
summary(casos_fil_covid_vacina)## DT_MIN VACINA_COV VACINA NU_IDADE_N
## Min. :2020-01-05 Min. :1.0 Min. :1.0 Min. : 1.00
## 1st Qu.:2020-11-05 1st Qu.:2.0 1st Qu.:2.0 1st Qu.:45.00
## Median :2021-03-21 Median :2.0 Median :2.0 Median :59.00
## Mean :2021-03-12 Mean :2.6 Mean :5.1 Mean :57.91
## 3rd Qu.:2021-06-10 3rd Qu.:2.0 3rd Qu.:9.0 3rd Qu.:72.00
## Max. :2022-12-04 Max. :9.0 Max. :9.0 Max. :99.00
## NA's :325989 NA's :526559
## EVOLUCAO
## Min. :1.0
## 1st Qu.:1.0
## Median :1.0
## Mean :1.5
## 3rd Qu.:2.0
## Max. :9.0
## NA's :99271Criando a coluna vacina cov e vacina gripe para entender se há um efeito de dupla vacinação positivo.
casos_fil_covid_vacina$DUPLA_VACINA <- if_else(casos_fil_covid_vacina$VACINA_COV == 1 & casos_fil_covid_vacina$VACINA == 1, 1 , 0)Estruturando em one hot encoded para poder fazer a regressão logit.
casos_fil_covid_vacina <- dummy_cols(casos_fil_covid_vacina, select_columns = c('VACINA_COV', 'VACINA','DUPLA_VACINA'),
remove_selected_columns = TRUE)
casos_fil_covid_vacina$EVOLUCAO <- if_else(casos_fil_covid_vacina$EVOLUCAO == 1 , 1, 0)Variáveis selecionadas: CS_SEX_F = se é mulher ou não, CS_RACA_2 = se é preto ou não, CS_ZONA_2 = se é rural ou não, NU_IDADE_N = idade da pessoa. EVOLUCAO = 1 para sobreviveu, 0 para óbito. Recordo também que o comando glm omite as linhas com valores nulos.
casos_fil_covid.logit <- glm(formula = EVOLUCAO ~ VACINA_1 + VACINA_COV_1 + DUPLA_VACINA_1+ NU_IDADE_N, family = binomial(link = "logit"),
data = casos_fil_covid_vacina)
summary(casos_fil_covid.logit)##
## Call:
## glm(formula = EVOLUCAO ~ VACINA_1 + VACINA_COV_1 + DUPLA_VACINA_1 +
## NU_IDADE_N, family = binomial(link = "logit"), data = casos_fil_covid_vacina)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.7098 -1.1571 0.6596 0.9119 1.6625
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 3.0445509 0.0074774 407.166 <2e-16 ***
## VACINA_1 0.3968219 0.0068614 57.834 <2e-16 ***
## VACINA_COV_1 0.3610720 0.0052273 69.074 <2e-16 ***
## DUPLA_VACINA_1 -0.1149745 0.0126732 -9.072 <2e-16 ***
## NU_IDADE_N -0.0417909 0.0001197 -349.236 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1755431 on 1364590 degrees of freedom
## Residual deviance: 1610875 on 1364586 degrees of freedom
## (767049 observations deleted due to missingness)
## AIC: 1610885
##
## Number of Fisher Scoring iterations: 4Análise dos sintomas
rm(casos_fil_covid_vacina)
gc()## used (Mb) gc trigger (Mb) max used (Mb)
## Ncells 5221079 278.9 12354841 659.9 5812999 310.5
## Vcells 510958394 3898.4 948513124 7236.6 948401121 7235.8casos_fil_covid_sintomas <- casos_fil_covid %>% select(DT_MIN, NU_IDADE_N, EVOLUCAO,FEBRE, TOSSE, GARGANTA, DISPNEIA, DESC_RESP, SATURACAO, DIARREIA, VOMITO, DOR_ABD, FADIGA, PERD_OLFT, PERD_PALA)
summary(casos_fil_covid_sintomas)## DT_MIN NU_IDADE_N EVOLUCAO FEBRE
## Min. :2020-01-05 Min. : 1.00 Min. :1.0 Min. :1.0
## 1st Qu.:2020-11-05 1st Qu.:45.00 1st Qu.:1.0 1st Qu.:1.0
## Median :2021-03-21 Median :59.00 Median :1.0 Median :1.0
## Mean :2021-03-12 Mean :57.91 Mean :1.5 Mean :1.4
## 3rd Qu.:2021-06-10 3rd Qu.:72.00 3rd Qu.:2.0 3rd Qu.:2.0
## Max. :2022-12-04 Max. :99.00 Max. :9.0 Max. :9.0
## NA's :99271 NA's :340351
## TOSSE GARGANTA DISPNEIA DESC_RESP
## Min. :1.0 Min. :1 Min. :1.00 Min. :1.0
## 1st Qu.:1.0 1st Qu.:2 1st Qu.:1.00 1st Qu.:1.0
## Median :1.0 Median :2 Median :1.00 Median :1.0
## Mean :1.3 Mean :2 Mean :1.28 Mean :1.4
## 3rd Qu.:1.0 3rd Qu.:2 3rd Qu.:1.00 3rd Qu.:2.0
## Max. :9.0 Max. :9 Max. :9.00 Max. :9.0
## NA's :268697 NA's :615677 NA's :265435 NA's :409211
## SATURACAO DIARREIA VOMITO DOR_ABD
## Min. :1.0 Min. :1 Min. :1.0 Min. :1.0
## 1st Qu.:1.0 1st Qu.:2 1st Qu.:2.0 1st Qu.:2.0
## Median :1.0 Median :2 Median :2.0 Median :2.0
## Mean :1.4 Mean :2 Mean :2.1 Mean :2.2
## 3rd Qu.:2.0 3rd Qu.:2 3rd Qu.:2.0 3rd Qu.:2.0
## Max. :9.0 Max. :9 Max. :9.0 Max. :9.0
## NA's :341842 NA's :643986 NA's :671957 NA's :873571
## FADIGA PERD_OLFT PERD_PALA
## Min. :1.0 Min. :1.0 Min. :1.0
## 1st Qu.:1.0 1st Qu.:2.0 1st Qu.:2.0
## Median :2.0 Median :2.0 Median :2.0
## Mean :1.9 Mean :2.1 Mean :2.1
## 3rd Qu.:2.0 3rd Qu.:2.0 3rd Qu.:2.0
## Max. :9.0 Max. :9.0 Max. :9.0
## NA's :796163 NA's :852465 NA's :853816Criando as dummies
casos_fil_covid_sintomas <- dummy_cols(casos_fil_covid_sintomas, select_columns = c("FEBRE" , "TOSSE", "GARGANTA", "DISPNEIA", "DESC_RESP", "SATURACAO", "DIARREIA", "VOMITO", "DOR_ABD", "FADIGA", "PERD_OLFT", "PERD_PALA" ),
remove_selected_columns = TRUE)
casos_fil_covid_sintomas$EVOLUCAO <- if_else(casos_fil_covid_sintomas$EVOLUCAO == 1 , 1, 0)Logit dos sintomas
casos_fil_covid.logit <- glm(formula = EVOLUCAO ~ FEBRE_1 + TOSSE_1 + GARGANTA_1 + DISPNEIA_1,
family = binomial(link = "logit"),
data = casos_fil_covid_sintomas)
summary(casos_fil_covid.logit)##
## Call:
## glm(formula = EVOLUCAO ~ FEBRE_1 + TOSSE_1 + GARGANTA_1 + DISPNEIA_1,
## family = binomial(link = "logit"), data = casos_fil_covid_sintomas)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.7954 -1.3919 0.8600 0.9164 1.1066
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.753840 0.004675 161.24 <2e-16 ***
## FEBRE_1 0.159034 0.003825 41.58 <2e-16 ***
## TOSSE_1 0.322382 0.004182 77.09 <2e-16 ***
## GARGANTA_1 0.153917 0.004632 33.23 <2e-16 ***
## DISPNEIA_1 -0.584917 0.004365 -134.00 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1786226 on 1394858 degrees of freedom
## Residual deviance: 1758084 on 1394854 degrees of freedom
## (736781 observations deleted due to missingness)
## AIC: 1758094
##
## Number of Fisher Scoring iterations: 4Tenho de separar em duas regressões, pois o R não imprime todos os valores, tem um limite a analisar.
casos_fil_covid.logit <- glm(formula = EVOLUCAO ~ DESC_RESP_1 + SATURACAO_1 + DIARREIA_1 + VOMITO_1 + DOR_ABD_1 + FADIGA_1 + PERD_OLFT_1 + PERD_PALA_1,
family = binomial(link = "logit"),
data = casos_fil_covid_sintomas)
summary(casos_fil_covid.logit)##
## Call:
## glm(formula = EVOLUCAO ~ DESC_RESP_1 + SATURACAO_1 + DIARREIA_1 +
## VOMITO_1 + DOR_ABD_1 + FADIGA_1 + PERD_OLFT_1 + PERD_PALA_1,
## family = binomial(link = "logit"), data = casos_fil_covid_sintomas)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.0685 -1.3002 0.7509 0.9405 1.0602
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.121971 0.004279 262.218 <2e-16 ***
## DESC_RESP_1 -0.375743 0.004563 -82.343 <2e-16 ***
## SATURACAO_1 -0.462214 0.004832 -95.653 <2e-16 ***
## DIARREIA_1 0.169312 0.006165 27.464 <2e-16 ***
## VOMITO_1 0.105631 0.007476 14.130 <2e-16 ***
## DOR_ABD_1 -0.002028 0.008309 -0.244 0.807
## FADIGA_1 0.133151 0.004603 28.929 <2e-16 ***
## PERD_OLFT_1 0.227462 0.010308 22.066 <2e-16 ***
## PERD_PALA_1 0.256625 0.010236 25.072 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1483778 on 1161598 degrees of freedom
## Residual deviance: 1451780 on 1161590 degrees of freedom
## (970041 observations deleted due to missingness)
## AIC: 1451798
##
## Number of Fisher Scoring iterations: 4Análise das comorbidades
rm(casos_fil_covid_sintomas)
gc()## used (Mb) gc trigger (Mb) max used (Mb)
## Ncells 5019034 268.1 12354841 659.9 5812999 310.5
## Vcells 553028984 4219.3 948513124 7236.6 948486590 7236.4casos_fil_covid_comorbidades <- casos_fil_covid %>% select(DT_MIN, NU_IDADE_N, EVOLUCAO, FATOR_RISC, PUERPERA, CARDIOPATI, HEMATOLOGI, SIND_DOWN, HEPATICA, ASMA, DIABETES, NEUROLOGIC, PNEUMOPATI, IMUNODEPRE, RENAL, OBESIDADE)
summary(casos_fil_covid_comorbidades)## DT_MIN NU_IDADE_N EVOLUCAO FATOR_RISC
## Min. :2020-01-05 Min. : 1.00 Min. :1.0 Length:2131640
## 1st Qu.:2020-11-05 1st Qu.:45.00 1st Qu.:1.0 Class :character
## Median :2021-03-21 Median :59.00 Median :1.0 Mode :character
## Mean :2021-03-12 Mean :57.91 Mean :1.5
## 3rd Qu.:2021-06-10 3rd Qu.:72.00 3rd Qu.:2.0
## Max. :2022-12-04 Max. :99.00 Max. :9.0
## NA's :99271
## PUERPERA CARDIOPATI HEMATOLOGI SIND_DOWN
## Min. :1.0 Min. :1.0 Min. :1.0 Min. :1.0
## 1st Qu.:2.0 1st Qu.:1.0 1st Qu.:2.0 1st Qu.:2.0
## Median :2.0 Median :1.0 Median :2.0 Median :2.0
## Mean :2.2 Mean :1.4 Mean :2.2 Mean :2.2
## 3rd Qu.:2.0 3rd Qu.:2.0 3rd Qu.:2.0 3rd Qu.:2.0
## Max. :9.0 Max. :9.0 Max. :9.0 Max. :9.0
## NA's :1352981 NA's :1099368 NA's :1350649 NA's :1353343
## HEPATICA ASMA DIABETES NEUROLOGIC
## Min. :1.0 Min. :1.0 Min. :1.0 Min. :1.0
## 1st Qu.:2.0 1st Qu.:2.0 1st Qu.:1.0 1st Qu.:2.0
## Median :2.0 Median :2.0 Median :2.0 Median :2.0
## Mean :2.2 Mean :2.1 Mean :1.6 Mean :2.1
## 3rd Qu.:2.0 3rd Qu.:2.0 3rd Qu.:2.0 3rd Qu.:2.0
## Max. :9.0 Max. :9.0 Max. :9.0 Max. :9.0
## NA's :1352960 NA's :1340898 NA's :1172542 NA's :1331091
## PNEUMOPATI IMUNODEPRE RENAL OBESIDADE
## Min. :1.0 Min. :1.0 Min. :1.0 Min. :1
## 1st Qu.:2.0 1st Qu.:2.0 1st Qu.:2.0 1st Qu.:2
## Median :2.0 Median :2.0 Median :2.0 Median :2
## Mean :2.1 Mean :2.1 Mean :2.1 Mean :2
## 3rd Qu.:2.0 3rd Qu.:2.0 3rd Qu.:2.0 3rd Qu.:2
## Max. :9.0 Max. :9.0 Max. :9.0 Max. :9
## NA's :1332442 NA's :1342756 NA's :1334452 NA's :1303592Criando as dummies
casos_fil_covid_comorbidades <- dummy_cols(casos_fil_covid_comorbidades, select_columns = c('FATOR_RISC', 'PUERPERA', 'CARDIOPATI', 'HEMATOLOGI', 'SIND_DOWN', 'HEPATICA', 'ASMA', 'DIABETES', 'NEUROLOGIC', 'PNEUMOPATI', 'IMUNODEPRE', 'RENAL', 'OBESIDADE'),
remove_selected_columns = TRUE)
casos_fil_covid_comorbidades$EVOLUCAO <- if_else(casos_fil_covid_comorbidades$EVOLUCAO == 1 , 1, 0)
summary(casos_fil_covid_comorbidades)## DT_MIN NU_IDADE_N EVOLUCAO FATOR_RISC_1
## Min. :2020-01-05 Min. : 1.00 Min. :0.00 Min. :0.0000
## 1st Qu.:2020-11-05 1st Qu.:45.00 1st Qu.:0.00 1st Qu.:0.0000
## Median :2021-03-21 Median :59.00 Median :1.00 Median :0.0000
## Mean :2021-03-12 Mean :57.91 Mean :0.65 Mean :0.3899
## 3rd Qu.:2021-06-10 3rd Qu.:72.00 3rd Qu.:1.00 3rd Qu.:1.0000
## Max. :2022-12-04 Max. :99.00 Max. :1.00 Max. :1.0000
## NA's :99271
## FATOR_RISC_2 FATOR_RISC_N FATOR_RISC_S PUERPERA_1
## Min. :0.0000 Min. :0.0000 Min. :0.0000 Min. :0
## 1st Qu.:0.0000 1st Qu.:0.0000 1st Qu.:0.0000 1st Qu.:0
## Median :0.0000 Median :0.0000 Median :0.0000 Median :0
## Mean :0.2743 Mean :0.1229 Mean :0.2128 Mean :0
## 3rd Qu.:1.0000 3rd Qu.:0.0000 3rd Qu.:0.0000 3rd Qu.:0
## Max. :1.0000 Max. :1.0000 Max. :1.0000 Max. :1
## NA's :1352981
## PUERPERA_2 PUERPERA_9 PUERPERA_NA CARDIOPATI_1
## Min. :0 Min. :0 Min. :0.0000 Min. :0.0
## 1st Qu.:1 1st Qu.:0 1st Qu.:0.0000 1st Qu.:0.0
## Median :1 Median :0 Median :1.0000 Median :1.0
## Mean :1 Mean :0 Mean :0.6347 Mean :0.6
## 3rd Qu.:1 3rd Qu.:0 3rd Qu.:1.0000 3rd Qu.:1.0
## Max. :1 Max. :1 Max. :1.0000 Max. :1.0
## NA's :1352981 NA's :1352981 NA's :1099368
## CARDIOPATI_2 CARDIOPATI_9 CARDIOPATI_NA HEMATOLOGI_1
## Min. :0.0 Min. :0 Min. :0.0000 Min. :0
## 1st Qu.:0.0 1st Qu.:0 1st Qu.:0.0000 1st Qu.:0
## Median :0.0 Median :0 Median :1.0000 Median :0
## Mean :0.3 Mean :0 Mean :0.5157 Mean :0
## 3rd Qu.:1.0 3rd Qu.:0 3rd Qu.:1.0000 3rd Qu.:0
## Max. :1.0 Max. :1 Max. :1.0000 Max. :1
## NA's :1099368 NA's :1099368 NA's :1350649
## HEMATOLOGI_2 HEMATOLOGI_9 HEMATOLOGI_NA SIND_DOWN_1
## Min. :0 Min. :0 Min. :0.0000 Min. :0
## 1st Qu.:1 1st Qu.:0 1st Qu.:0.0000 1st Qu.:0
## Median :1 Median :0 Median :1.0000 Median :0
## Mean :1 Mean :0 Mean :0.6336 Mean :0
## 3rd Qu.:1 3rd Qu.:0 3rd Qu.:1.0000 3rd Qu.:0
## Max. :1 Max. :1 Max. :1.0000 Max. :1
## NA's :1350649 NA's :1350649 NA's :1353343
## SIND_DOWN_2 SIND_DOWN_9 SIND_DOWN_NA HEPATICA_1
## Min. :0 Min. :0 Min. :0.0000 Min. :0
## 1st Qu.:1 1st Qu.:0 1st Qu.:0.0000 1st Qu.:0
## Median :1 Median :0 Median :1.0000 Median :0
## Mean :1 Mean :0 Mean :0.6349 Mean :0
## 3rd Qu.:1 3rd Qu.:0 3rd Qu.:1.0000 3rd Qu.:0
## Max. :1 Max. :1 Max. :1.0000 Max. :1
## NA's :1353343 NA's :1353343 NA's :1352960
## HEPATICA_2 HEPATICA_9 HEPATICA_NA ASMA_1
## Min. :0 Min. :0 Min. :0.0000 Min. :0.0
## 1st Qu.:1 1st Qu.:0 1st Qu.:0.0000 1st Qu.:0.0
## Median :1 Median :0 Median :1.0000 Median :0.0
## Mean :1 Mean :0 Mean :0.6347 Mean :0.1
## 3rd Qu.:1 3rd Qu.:0 3rd Qu.:1.0000 3rd Qu.:0.0
## Max. :1 Max. :1 Max. :1.0000 Max. :1.0
## NA's :1352960 NA's :1352960 NA's :1340898
## ASMA_2 ASMA_9 ASMA_NA DIABETES_1
## Min. :0.0 Min. :0 Min. :0.000 Min. :0.0
## 1st Qu.:1.0 1st Qu.:0 1st Qu.:0.000 1st Qu.:0.0
## Median :1.0 Median :0 Median :1.000 Median :0.0
## Mean :0.9 Mean :0 Mean :0.629 Mean :0.5
## 3rd Qu.:1.0 3rd Qu.:0 3rd Qu.:1.000 3rd Qu.:1.0
## Max. :1.0 Max. :1 Max. :1.000 Max. :1.0
## NA's :1340898 NA's :1340898 NA's :1172542
## DIABETES_2 DIABETES_9 DIABETES_NA NEUROLOGIC_1
## Min. :0.0 Min. :0 Min. :0.0000 Min. :0.0
## 1st Qu.:0.0 1st Qu.:0 1st Qu.:0.0000 1st Qu.:0.0
## Median :0.0 Median :0 Median :1.0000 Median :0.0
## Mean :0.5 Mean :0 Mean :0.5501 Mean :0.1
## 3rd Qu.:1.0 3rd Qu.:0 3rd Qu.:1.0000 3rd Qu.:0.0
## Max. :1.0 Max. :1 Max. :1.0000 Max. :1.0
## NA's :1172542 NA's :1172542 NA's :1331091
## NEUROLOGIC_2 NEUROLOGIC_9 NEUROLOGIC_NA PNEUMOPATI_1
## Min. :0.0 Min. :0 Min. :0.0000 Min. :0.0
## 1st Qu.:1.0 1st Qu.:0 1st Qu.:0.0000 1st Qu.:0.0
## Median :1.0 Median :0 Median :1.0000 Median :0.0
## Mean :0.9 Mean :0 Mean :0.6244 Mean :0.1
## 3rd Qu.:1.0 3rd Qu.:0 3rd Qu.:1.0000 3rd Qu.:0.0
## Max. :1.0 Max. :1 Max. :1.0000 Max. :1.0
## NA's :1331091 NA's :1331091 NA's :1332442
## PNEUMOPATI_2 PNEUMOPATI_9 PNEUMOPATI_NA IMUNODEPRE_1
## Min. :0.0 Min. :0 Min. :0.0000 Min. :0.0
## 1st Qu.:1.0 1st Qu.:0 1st Qu.:0.0000 1st Qu.:0.0
## Median :1.0 Median :0 Median :1.0000 Median :0.0
## Mean :0.9 Mean :0 Mean :0.6251 Mean :0.1
## 3rd Qu.:1.0 3rd Qu.:0 3rd Qu.:1.0000 3rd Qu.:0.0
## Max. :1.0 Max. :1 Max. :1.0000 Max. :1.0
## NA's :1332442 NA's :1332442 NA's :1342756
## IMUNODEPRE_2 IMUNODEPRE_9 IMUNODEPRE_NA RENAL_1
## Min. :0.0 Min. :0 Min. :0.0000 Min. :0.0
## 1st Qu.:1.0 1st Qu.:0 1st Qu.:0.0000 1st Qu.:0.0
## Median :1.0 Median :0 Median :1.0000 Median :0.0
## Mean :0.9 Mean :0 Mean :0.6299 Mean :0.1
## 3rd Qu.:1.0 3rd Qu.:0 3rd Qu.:1.0000 3rd Qu.:0.0
## Max. :1.0 Max. :1 Max. :1.0000 Max. :1.0
## NA's :1342756 NA's :1342756 NA's :1334452
## RENAL_2 RENAL_9 RENAL_NA OBESIDADE_1
## Min. :0.0 Min. :0 Min. :0.000 Min. :0.0
## 1st Qu.:1.0 1st Qu.:0 1st Qu.:0.000 1st Qu.:0.0
## Median :1.0 Median :0 Median :1.000 Median :0.0
## Mean :0.9 Mean :0 Mean :0.626 Mean :0.2
## 3rd Qu.:1.0 3rd Qu.:0 3rd Qu.:1.000 3rd Qu.:0.0
## Max. :1.0 Max. :1 Max. :1.000 Max. :1.0
## NA's :1334452 NA's :1334452 NA's :1303592
## OBESIDADE_2 OBESIDADE_9 OBESIDADE_NA
## Min. :0.0 Min. :0 Min. :0.0000
## 1st Qu.:1.0 1st Qu.:0 1st Qu.:0.0000
## Median :1.0 Median :0 Median :1.0000
## Mean :0.8 Mean :0 Mean :0.6115
## 3rd Qu.:1.0 3rd Qu.:0 3rd Qu.:1.0000
## Max. :1.0 Max. :1 Max. :1.0000
## NA's :1303592 NA's :1303592Logit comorbidades
casos_fil_covid.logit <- glm(formula = EVOLUCAO ~ FATOR_RISC_1 + PUERPERA_1 + CARDIOPATI_1 + HEMATOLOGI_1 + SIND_DOWN_1 + HEPATICA_1 + ASMA_1 + DIABETES_1 + NEUROLOGIC_1 + PNEUMOPATI_1 + IMUNODEPRE_1 + RENAL_1 + OBESIDADE_1 ,
family = binomial(link = "logit"),
data = casos_fil_covid_comorbidades)
summary(casos_fil_covid.logit)##
## Call:
## glm(formula = EVOLUCAO ~ FATOR_RISC_1 + PUERPERA_1 + CARDIOPATI_1 +
## HEMATOLOGI_1 + SIND_DOWN_1 + HEPATICA_1 + ASMA_1 + DIABETES_1 +
## NEUROLOGIC_1 + PNEUMOPATI_1 + IMUNODEPRE_1 + RENAL_1 + OBESIDADE_1,
## family = binomial(link = "logit"), data = casos_fil_covid_comorbidades)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.0496 -1.3077 0.8999 0.9911 2.2419
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.761552 0.005652 134.747 < 2e-16 ***
## FATOR_RISC_1 -0.066832 0.005127 -13.035 < 2e-16 ***
## PUERPERA_1 0.777301 0.041216 18.859 < 2e-16 ***
## CARDIOPATI_1 -0.221110 0.004977 -44.429 < 2e-16 ***
## HEMATOLOGI_1 -0.110694 0.022751 -4.865 1.14e-06 ***
## SIND_DOWN_1 -0.002804 0.034751 -0.081 0.936
## HEPATICA_1 -0.447586 0.021010 -21.304 < 2e-16 ***
## ASMA_1 0.431097 0.013041 33.056 < 2e-16 ***
## DIABETES_1 -0.239192 0.005109 -46.820 < 2e-16 ***
## NEUROLOGIC_1 -0.564989 0.009985 -56.585 < 2e-16 ***
## PNEUMOPATI_1 -0.502164 0.010495 -47.850 < 2e-16 ***
## IMUNODEPRE_1 -0.404165 0.012380 -32.647 < 2e-16 ***
## RENAL_1 -0.591436 0.010353 -57.130 < 2e-16 ***
## OBESIDADE_1 -0.041856 0.007138 -5.863 4.53e-09 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 954775 on 708893 degrees of freedom
## Residual deviance: 937544 on 708880 degrees of freedom
## (1422746 observations deleted due to missingness)
## AIC: 937572
##
## Number of Fisher Scoring iterations: 4Descobrimos agora quais variáveis sãos impactantes para ato da chance de a pessoa internada vir a óbito, portanto vamos comparar essas variáveis ao longo do tempo, inicialmente regredindo o logit com elas em dois grupos, pré e pós vacina.(17/01/2021)
As variáveis são, Demográficas = idade e raça. Vacina = Vacina covid e vacina gripe. Sintomas = dispinéia, desconforto respiratírio e saturação, Se foi para UTI, Comorbidades = hepatica, pneumopatia, neurologica, renal e cardiopatia. Entretanto como as comorbidades me fazem perder muitas linhas por nulos vou desconsidera-las por enquanto.
rm(casos_fil_covid_comorbidades)
gc()## used (Mb) gc trigger (Mb) max used (Mb)
## Ncells 4566641 243.9 12354841 659.9 5812999 310.5
## Vcells 551752373 4209.6 948513124 7236.6 948486590 7236.4casos_fil_covid_analise_vacina <- casos_fil_covid %>% select(DT_MIN, NU_IDADE_N, CS_RACA, VACINA, VACINA_COV, UTI, DISPNEIA, DESC_RESP,SATURACAO, EVOLUCAO )
summary(casos_fil_covid_analise_vacina)## DT_MIN NU_IDADE_N CS_RACA VACINA
## Min. :2020-01-05 Min. : 1.00 Min. :1.00 Min. :1.0
## 1st Qu.:2020-11-05 1st Qu.:45.00 1st Qu.:1.00 1st Qu.:2.0
## Median :2021-03-21 Median :59.00 Median :4.00 Median :2.0
## Mean :2021-03-12 Mean :57.91 Mean :3.49 Mean :5.1
## 3rd Qu.:2021-06-10 3rd Qu.:72.00 3rd Qu.:4.00 3rd Qu.:9.0
## Max. :2022-12-04 Max. :99.00 Max. :9.00 Max. :9.0
## NA's :28517 NA's :526559
## VACINA_COV UTI DISPNEIA DESC_RESP
## Min. :1.0 Min. :1.00 Min. :1.00 Min. :1.0
## 1st Qu.:2.0 1st Qu.:1.00 1st Qu.:1.00 1st Qu.:1.0
## Median :2.0 Median :2.00 Median :1.00 Median :1.0
## Mean :2.6 Mean :1.79 Mean :1.28 Mean :1.4
## 3rd Qu.:2.0 3rd Qu.:2.00 3rd Qu.:1.00 3rd Qu.:2.0
## Max. :9.0 Max. :9.00 Max. :9.00 Max. :9.0
## NA's :325989 NA's :253860 NA's :265435 NA's :409211
## SATURACAO EVOLUCAO
## Min. :1.0 Min. :1.0
## 1st Qu.:1.0 1st Qu.:1.0
## Median :1.0 Median :1.0
## Mean :1.4 Mean :1.5
## 3rd Qu.:2.0 3rd Qu.:2.0
## Max. :9.0 Max. :9.0
## NA's :341842 NA's :99271Dummies
casos_fil_covid_analise_vacina <- dummy_cols(casos_fil_covid_analise_vacina, select_columns = c('CS_RACA', 'VACINA', 'VACINA_COV','UTI', 'DISPNEIA', 'DESC_RESP', 'SATURACAO'),
remove_selected_columns = TRUE)
casos_fil_covid_analise_vacina$EVOLUCAO <- if_else(casos_fil_covid_analise_vacina$EVOLUCAO == 1 , 1, 0)Separando em grupos pré e pós vacinação
casos_fil_covid_analise_vacina_pre <- casos_fil_covid_analise_vacina %>% filter(DT_MIN < '2021-01-01')
casos_fil_covid_analise_vacina_pos <- casos_fil_covid_analise_vacina %>% filter(DT_MIN >= '2021-01-01')Logit pré vacina
casos_fil_covid.logit <- glm(formula = EVOLUCAO ~ NU_IDADE_N + CS_RACA_2 + VACINA_1 + UTI_1 + DISPNEIA_1 + DESC_RESP_1 + SATURACAO_1,
family = binomial(link = "logit"),
data = casos_fil_covid_analise_vacina_pre)
summary(casos_fil_covid.logit)##
## Call:
## glm(formula = EVOLUCAO ~ NU_IDADE_N + CS_RACA_2 + VACINA_1 +
## UTI_1 + DISPNEIA_1 + DESC_RESP_1 + SATURACAO_1, family = binomial(link = "logit"),
## data = casos_fil_covid_analise_vacina_pre)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -3.1604 -0.9013 0.4780 0.8014 2.2124
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 4.7307670 0.0190467 248.38 <2e-16 ***
## NU_IDADE_N -0.0465783 0.0002466 -188.88 <2e-16 ***
## CS_RACA_2 -0.3150523 0.0168036 -18.75 <2e-16 ***
## VACINA_1 0.3029572 0.0102442 29.57 <2e-16 ***
## UTI_1 -1.4521395 0.0076773 -189.15 <2e-16 ***
## DISPNEIA_1 -0.2522193 0.0100036 -25.21 <2e-16 ***
## DESC_RESP_1 -0.3289754 0.0090848 -36.21 <2e-16 ***
## SATURACAO_1 -0.3142151 0.0091357 -34.39 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 521846 on 404498 degrees of freedom
## Residual deviance: 420593 on 404491 degrees of freedom
## (278120 observations deleted due to missingness)
## AIC: 420609
##
## Number of Fisher Scoring iterations: 4Logit pós vacina. Parece que não há diferença entre as duas regressões, o que indica que os grupos são semelhantes, o que nos permite utilizar toda a base de dados para uma unica regressão, não precisando separar período pós e pré vacina.
casos_fil_covid.logit <- glm(formula = EVOLUCAO ~ NU_IDADE_N + CS_RACA_2 + VACINA_1 + VACINA_COV_1 + UTI_1 + DISPNEIA_1 + DESC_RESP_1 + SATURACAO_1,
family = binomial(link = "logit"),
data = casos_fil_covid_analise_vacina_pos)
summary(casos_fil_covid.logit)##
## Call:
## glm(formula = EVOLUCAO ~ NU_IDADE_N + CS_RACA_2 + VACINA_1 +
## VACINA_COV_1 + UTI_1 + DISPNEIA_1 + DESC_RESP_1 + SATURACAO_1,
## family = binomial(link = "logit"), data = casos_fil_covid_analise_vacina_pos)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.9564 -0.9074 0.5129 0.7872 2.2449
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 4.1255197 0.0138877 297.06 <2e-16 ***
## NU_IDADE_N -0.0413358 0.0001874 -220.56 <2e-16 ***
## CS_RACA_2 -0.2732955 0.0139572 -19.58 <2e-16 ***
## VACINA_1 0.1920415 0.0093885 20.45 <2e-16 ***
## VACINA_COV_1 0.4385190 0.0066027 66.42 <2e-16 ***
## UTI_1 -1.6605966 0.0059225 -280.39 <2e-16 ***
## DISPNEIA_1 -0.1945834 0.0079570 -24.45 <2e-16 ***
## DESC_RESP_1 -0.2774648 0.0070134 -39.56 <2e-16 ***
## SATURACAO_1 -0.2285727 0.0077752 -29.40 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 872348 on 677486 degrees of freedom
## Residual deviance: 713446 on 677478 degrees of freedom
## (771534 observations deleted due to missingness)
## AIC: 713464
##
## Number of Fisher Scoring iterations: 4Criando coluna idade, cortando número de idades iguais ou menores do que zero “erro de input” e excluindo estes valores.
rm(casos_fil_covid_analise_vacina,casos_fil_covid_analise_vacina_pre,casos_fil_covid_analise_vacina_pos)
gc()## used (Mb) gc trigger (Mb) max used (Mb)
## Ncells 4535227 242.3 12354841 659.9 5812999 310.5
## Vcells 510324175 3893.5 950007366 7248.0 950007366 7248.0casos_fil_covid$DT_NASC<- as_date(casos_fil_covid$DT_NASC, format= "%d/%m/%Y")
casos_fil_covid$idade <- (as.numeric(casos_fil_covid$DT_MIN) - as.numeric(casos_fil_covid$DT_NASC))/365
count(casos_fil_covid, idade <= 0)## # A tibble: 3 x 2
## `idade <= 0` n
## <lgl> <int>
## 1 FALSE 2129657
## 2 TRUE 144
## 3 NA 1839count(casos_fil_covid, idade >= 100)## # A tibble: 3 x 2
## `idade >= 100` n
## <lgl> <int>
## 1 FALSE 2129663
## 2 TRUE 138
## 3 NA 1839casos_fil_covid <- filter(casos_fil_covid, idade > 0)
casos_fil_covid <- filter(casos_fil_covid, idade < 100)Devemos fazer mais umas análises sobre as bases para poder cravarmos se possuem uma distribuição similar ou não. vamos olhar parqa a caracteristica das pessoas que foram à óbito. Linha azul representa início da vacinação.
a <- casos_fil_covid %>% filter(EVOLUCAO == 2) %>% group_by(SEM_DTMIN) %>% summarise(avg = mean(NU_IDADE_N))
ggplot(data = a, aes(x = SEM_DTMIN, y = avg )) + geom_line() + ggtitle("Média de idade dos óbitos ao longo das semanas epidemiológicas") + geom_vline(xintercept=55,lwd=1,colour="blue",linetype = 'dotted')## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## i Please use `linewidth` instead.
Distribuição das raças, pode parecer a priori que a mortalidade dos pardos no inicio era maior, mas o que parece ter ocorrido é que no começo os brancos estavam também sido deixados como nulos.
a <- casos_fil_covid %>% filter(EVOLUCAO == 2) %>% group_by(SEM_DTMIN) %>% count(CS_RACA)
a$CS_RACA <- replace(a$CS_RACA,a$CS_RACA == 1,"Branca")
a$CS_RACA <- replace(a$CS_RACA,a$CS_RACA == 2,"Preta")
a$CS_RACA <- replace(a$CS_RACA,a$CS_RACA == 3,"Amarela")
a$CS_RACA <- replace(a$CS_RACA,a$CS_RACA == 4,"Parda")
a$CS_RACA <- replace(a$CS_RACA,a$CS_RACA == 5,"Indigena")
a$CS_RACA <- replace(a$CS_RACA,a$CS_RACA == 9,"Ignorado")
ggplot(data = a, aes(x = SEM_DTMIN, y = n, colour = CS_RACA)) + geom_line() + ggtitle("Óbitos por raça ao longo das semanas") + geom_vline(xintercept=55,lwd=1,colour="blue",linetype = 'dotted')
Vacina covid, como esperado os mortos vacinados se tornam maioria ao longo do tempo, mas como sabemos que a vacina possui efeito de aumentar a chance de recuperação podemos dizer que é só porque os vacinados se tornam maioria.
a <- casos_fil_covid %>% filter(EVOLUCAO == 2) %>% group_by(SEM_DTMIN) %>% count(VACINA_COV)
a$VACINA_COV <- replace(a$VACINA_COV,a$VACINA_COV == 1,"Vacinados")
a$VACINA_COV <- replace(a$VACINA_COV,a$VACINA_COV == 2,"Não Vacinados")
a$VACINA_COV <- replace(a$VACINA_COV,a$VACINA_COV == 9,"Ignorado")
ggplot(data = a, aes(x = SEM_DTMIN, y = n, colour = VACINA_COV)) + geom_line() + ggtitle("Número de óbitos entre vacinados") + geom_vline(xintercept=55,lwd=1,colour="blue",linetype = 'dotted')
Vacina gripe
a <- casos_fil_covid %>% filter(EVOLUCAO == 2) %>% group_by(SEM_DTMIN) %>% count(VACINA)
a$VACINA <- replace(a$VACINA,a$VACINA == 1,"Vacinados")
a$VACINA <- replace(a$VACINA,a$VACINA == 2,"Não Vacinados")
a$VACINA <- replace(a$VACINA,a$VACINA == 9,"Ignorado")
ggplot(data = a, aes(x = SEM_DTMIN, y = n, colour = VACINA)) + geom_line() + ggtitle("Número de óbitos entre vacinados contra gripe") + geom_vline(xintercept=55,lwd=1,colour="blue",linetype = 'dotted')
Dispnéia, aparece estar em paralelo com os casos do covid
a <- casos_fil_covid %>% filter(EVOLUCAO == 2) %>% group_by(SEM_DTMIN) %>% count(DESC_RESP)
a$DESC_RESP <- replace(a$DESC_RESP,a$DESC_RESP == 1,"Sim")
a$DESC_RESP <- replace(a$DESC_RESP,a$DESC_RESP == 2,"Não")
a$DESC_RESP <- replace(a$DESC_RESP,a$DESC_RESP == 9,"Ignorado")
ggplot(data = a, aes(x = SEM_DTMIN, y = n, colour = DESC_RESP)) + geom_line() + ggtitle("Número de óbitos de pessoas com desconforto respiratório") + geom_vline(xintercept=55,lwd=1,colour="blue",linetype = 'dotted')
Desconforto respiratório, nada diferente da dispinéia.
a <- casos_fil_covid %>% filter(EVOLUCAO == 2) %>% group_by(SEM_DTMIN) %>% count(DISPNEIA)
a$DISPNEIA <- replace(a$DISPNEIA,a$DISPNEIA == 1,"Sim")
a$DISPNEIA <- replace(a$DISPNEIA,a$DISPNEIA == 2,"Não")
a$DISPNEIA <- replace(a$DISPNEIA,a$DISPNEIA == 9,"Ignorado")
ggplot(data = a, aes(x = SEM_DTMIN, y = n, colour = DISPNEIA)) + geom_line() + ggtitle("Número de óbitos de pessoas com dispinéia") + geom_vline(xintercept=55,lwd=1,colour="blue",linetype = 'dotted')
Saturação, nada diferente da dispinéia.
a <- casos_fil_covid %>% filter(EVOLUCAO == 2) %>% group_by(SEM_DTMIN) %>% count(SATURACAO)
a$SATURACAO <- replace(a$SATURACAO,a$SATURACAO == 1,"Sim")
a$SATURACAO <- replace(a$SATURACAO,a$SATURACAO == 2,"Não")
a$SATURACAO <- replace(a$SATURACAO,a$SATURACAO == 9,"Ignorado")
ggplot(data = a, aes(x = SEM_DTMIN, y = n, colour = SATURACAO)) + geom_line() + ggtitle("Número de óbitos de pessoas com dispinéia") + geom_vline(xintercept=55,lwd=1,colour="blue",linetype = 'dotted')
UTI
a <- casos_fil_covid %>% filter(EVOLUCAO == 2) %>% group_by(SEM_DTMIN) %>% count(UTI)
a$UTI <- replace(a$UTI,a$UTI == 1,"Sim")
a$UTI <- replace(a$UTI,a$UTI == 2,"Não")
a$UTI <- replace(a$UTI,a$UTI == 9,"Ignorado")
ggplot(data = a, aes(x = SEM_DTMIN, y = n, colour = UTI)) + geom_line() + ggtitle("Número de óbitos de pessoas que foram para UTI") + geom_vline(xintercept=55,lwd=1,colour="blue",linetype = 'dotted')
Hepática, numero de nulos alto.
a <- casos_fil_covid %>% filter(EVOLUCAO == 2) %>% group_by(SEM_DTMIN) %>% count(HEPATICA)
a$HEPATICA <- replace(a$HEPATICA,a$HEPATICA == 1,"Sim")
a$HEPATICA <- replace(a$HEPATICA,a$HEPATICA == 2,"Não")
a$HEPATICA <- replace(a$HEPATICA,a$HEPATICA == 9,"Ignorado")
ggplot(data = a, aes(x = SEM_DTMIN, y = n, colour = HEPATICA)) + geom_line() + ggtitle("Número de óbitos de pessoas hepáticas") + geom_vline(xintercept=55,lwd=1,colour="blue",linetype = 'dotted')
Pneupática, numero de nulos alto.
a <- casos_fil_covid %>% filter(EVOLUCAO == 2) %>% group_by(SEM_DTMIN) %>% count(PNEUMOPATI)
a$PNEUMOPATI <- replace(a$PNEUMOPATI,a$PNEUMOPATI == 1,"Sim")
a$PNEUMOPATI <- replace(a$PNEUMOPATI,a$PNEUMOPATI == 2,"Não")
a$PNEUMOPATI <- replace(a$PNEUMOPATI,a$PNEUMOPATI == 9,"Ignorado")
ggplot(data = a, aes(x = SEM_DTMIN, y = n, colour = PNEUMOPATI)) + geom_line() + ggtitle("Número de óbitos de pessoas pneupáticas") + geom_vline(xintercept=55,lwd=1,colour="blue",linetype = 'dotted')
### Neurológica, numero de nulos alto.
a <- casos_fil_covid %>% filter(EVOLUCAO == 2) %>% group_by(SEM_DTMIN) %>% count(NEUROLOGIC)
a$NEUROLOGIC <- replace(a$NEUROLOGIC,a$NEUROLOGIC == 1,"Sim")
a$NEUROLOGIC <- replace(a$NEUROLOGIC,a$NEUROLOGIC == 2,"Não")
a$NEUROLOGIC <- replace(a$NEUROLOGIC,a$NEUROLOGIC == 9,"Ignorado")
ggplot(data = a, aes(x = SEM_DTMIN, y = n, colour = NEUROLOGIC)) + geom_line() + ggtitle("Número de óbitos de pessoas pneupáticas") + geom_vline(xintercept=55,lwd=1,colour="blue",linetype = 'dotted')
Renal, numero de nulos alto.
a <- casos_fil_covid %>% filter(EVOLUCAO == 2) %>% group_by(SEM_DTMIN) %>% count(RENAL)
c <- colnames(a)[2]
colnames(a)[2] = 'b'
a$b <- replace(a$b,a$b == 1,"Sim")
a$b <- replace(a$b,a$b == 2,"Não")
a$b <- replace(a$b,a$b == 9,"Ignorado")
colnames(a)[2] = c
ggplot(data = a, aes(x = SEM_DTMIN, y = n, colour = RENAL)) + geom_line() + ggtitle("Número de óbitos de pessoas com problema renal") + geom_vline(xintercept=55,lwd=1,colour="blue",linetype = 'dotted')
Cardiopatia, número de nulos alto.
a <- casos_fil_covid %>% filter(EVOLUCAO == 2) %>% group_by(SEM_DTMIN) %>% count(CARDIOPATI)
a$CARDIOPATI <- replace(a$CARDIOPATI,a$CARDIOPATI == 1,"Sim")
a$CARDIOPATI <- replace(a$CARDIOPATI,a$CARDIOPATI == 2,"Não")
a$CARDIOPATI <- replace(a$CARDIOPATI,a$CARDIOPATI == 9,"Ignorado")
ggplot(data = a, aes(x = SEM_DTMIN, y = n, colour = CARDIOPATI)) + geom_line() + ggtitle("Número de óbitos de pessoas pneupáticas") + geom_vline(xintercept=55,lwd=1,colour="blue",linetype = 'dotted')