BMA- Health belief model
Binh Thang
2022-10-03
Theoretical concept for study
library(foreign)
data <- read.csv("C:/Users/binht/Dropbox/Hue/data/2022_data_consultant/01 DD 1/Rcode/data2_1.csv")- BMA model
library(BMA)## Loading required package: survival## Loading required package: leaps## Loading required package: robustbase##
## Attaching package: 'robustbase'## The following object is masked from 'package:survival':
##
## heart## Loading required package: inline## Loading required package: rrcov## Scalable Robust Estimators with High Breakdown Point (version 1.5-5)names(data)## [1] "TAI2group" "TONG_LYTHUYET" "C92Caregiver_spouse"
## [4] "C17Smoking_present" "TONG_KIENTHUCCHUNG" "PL_THUCHANH"
## [7] "TONGmMRC" "TONGCAT" "TONGYDINH"data$C17Smoking_present=as.factor(data$C17Smoking_present)
data$PL_THUCHANH=as.factor(data$PL_THUCHANH)
data$C92Caregiver_spouse=as.factor(data$C92Caregiver_spouse)y=data[,1]
x=data[ ,-1]bma =bic.glm(x, y, strict = FALSE, OR = 20,data=data, glm.family="binomial", factor.type=TRUE)summary(bma)##
## Call:
## bic.glm.data.frame(x = x, y = y, glm.family = "binomial", strict = FALSE, OR = 20, factor.type = TRUE, data = data)
##
##
## 6 models were selected
## Best 5 models (cumulative posterior probability = 0.9593 ):
##
## p!=0 EV SD model 1 model 2
## Intercept 100 -14.803951 2.11590 -15.0460 -13.2882
## TONG_LYTHUYET 100.0 0.323197 0.05619 0.3173 0.3236
## C92Caregiver_spouse 4.1
## .1 -0.020066 0.23310 . .
## C17Smoking_present 4.2
## .1 -0.021348 0.22611 . .
## TONG_KIENTHUCCHUNG 0.0 0.000000 0.00000 . .
## PL_THUCHANH 4.9
## .1 0.037602 0.27816 . .
## TONGmMRC 0.0 0.000000 0.00000 . .
## TONGCAT 6.7 -0.009056 0.04509 . -0.1349
## TONGYDINH 5.6 -0.105286 0.65001 . .
##
## nVar 1 2
## BIC -2485.0781 -2480.2610
## post prob 0.746 0.067
## model 3 model 4 model 5
## Intercept -13.2706 -15.1328 -14.7143
## TONG_LYTHUYET 0.4212 0.3127 0.3154
## C92Caregiver_spouse
## .1 . . .
## C17Smoking_present
## .1 . . -0.5125
## TONG_KIENTHUCCHUNG . . .
## PL_THUCHANH
## .1 . 0.7733 .
## TONGmMRC . . .
## TONGCAT . . .
## TONGYDINH -1.8857 . .
##
## nVar 2 2 2
## BIC -2479.8930 -2479.6165 -2479.3073
## post prob 0.056 0.049 0.042a=imageplot.bma(bma)
library("epiDisplay")## Loading required package: MASS## Loading required package: nnetm1= glm(TAI2group ~ TONGYDINH, family=binomial, data=data)
logistic.display(m1)##
## Logistic regression predicting TAI2group
##
## OR(95%CI) P(Wald's test) P(LR-test)
## TONGYDINH (cont. var.) 814.06 (161.34,4107.51) < 0.001 < 0.001
##
## Log-likelihood = -26.1884
## No. of observations = 420
## AIC value = 56.3769m2= glm(TAI2group ~ TONGYDINH+PL_THUCHANH, family=binomial, data=data)
logistic.display(m2)##
## Logistic regression predicting TAI2group
##
## crude OR(95%CI) adj. OR(95%CI)
## TONGYDINH (cont. var.) 814.06 (161.34,4107.51) 696.6 (137.78,3521.82)
##
## PL_THUCHANH: 1 vs 0 3.13 (2.06,4.75) 2.29 (0.43,12.32)
##
## P(Wald's test) P(LR-test)
## TONGYDINH (cont. var.) < 0.001 < 0.001
##
## PL_THUCHANH: 1 vs 0 0.335 0.339
##
## Log-likelihood = -25.7321
## No. of observations = 420
## AIC value = 57.4643m3= glm(TAI2group ~ TONGYDINH+C17Smoking_present , family=binomial, data=data)
logistic.display(m3)##
## Logistic regression predicting TAI2group
##
## crude OR(95%CI) adj. OR(95%CI)
## TONGYDINH (cont. var.) 814.06 (161.34,4107.51) 787.34 (156,3973.79)
##
## C17Smoking_present: 1 vs 0 0.65 (0.44,0.98) 0.69 (0.13,3.54)
##
## P(Wald's test) P(LR-test)
## TONGYDINH (cont. var.) < 0.001 < 0.001
##
## C17Smoking_present: 1 vs 0 0.657 0.658
##
## Log-likelihood = -26.0903
## No. of observations = 420
## AIC value = 58.1805m4= glm(TAI2group ~ TONGYDINH+C92Caregiver_spouse, family=binomial, data=data)
logistic.display(m4)##
## Logistic regression predicting TAI2group
##
## crude OR(95%CI) adj. OR(95%CI)
## TONGYDINH (cont. var.) 814.06 (161.34,4107.51) 789.43 (156.41,3984.33)
##
## C92Caregiver_spouse: 1 vs 0 0.63 (0.41,0.97) 0.68 (0.11,4.07)
##
## P(Wald's test) P(LR-test)
## TONGYDINH (cont. var.) < 0.001 < 0.001
##
## C92Caregiver_spouse: 1 vs 0 0.671 0.671
##
## Log-likelihood = -26.0981
## No. of observations = 420
## AIC value = 58.1962m5= glm(TAI2group ~ TONGYDINH+TONGmMRC, family=binomial, data=data)
logistic.display(m5)##
## Logistic regression predicting TAI2group
##
## crude OR(95%CI) adj. OR(95%CI)
## TONGYDINH (cont. var.) 814.06 (161.34,4107.51) 800.48 (149.37,4289.97)
##
## TONGmMRC (cont. var.) 0.55 (0.39,0.76) 0.95 (0.25,3.68)
##
## P(Wald's test) P(LR-test)
## TONGYDINH (cont. var.) < 0.001 < 0.001
##
## TONGmMRC (cont. var.) 0.944 0.944
##
## Log-likelihood = -26.186
## No. of observations = 420
## AIC value = 58.3719