3.3 SEP, Education, HIV Prevalence

We turn our attention now to the association between these two measures of SEP and HIV prevalence.

Let’s conduct some initial analyses:

#--- Link PC1 & PC1 quintiles back into the tz dataset
tz <- assets %>% select(ID, pc1, pc1.q) %>% left_join(tz, ., by = "ID")

#--- Explicitly set the baseline of serostat to be "hiv negative"
tz <- tz %>% mutate(serostat = fct_relevel(serostat, "hiv negative"))

#--- Is there an association between Asset Index & HIV Prevalence?
tz %$% tabpct(serostat, pc1.q, percent = "col", graph = F)

## 
## Column percent 
##                pc1.q
## serostat         Low SEP       %  Low-Middle SEP       %  Middle SEP
##   hiv negative       532  (96.9)             533  (97.3)         537
##   hiv  positive       17   (3.1)              15   (2.7)          11
##   Total              549   (100)             548   (100)         548
##                pc1.q
## serostat             %  High-Middle SEP       %  High SEP       %
##   hiv negative    (98)              540  (98.5)       541  (98.7)
##   hiv  positive    (2)                8   (1.5)         7   (1.3)
##   Total          (100)              548   (100)       548   (100)

tz %$% cc(serostat, pc1.q, graph = F)

## 
##                 pc1.q
## serostat         Low SEP Low-Middle SEP Middle SEP High-Middle SEP
##   hiv negative   532     533            537        540            
##   hiv  positive  17      15             11         8              
##                                                                   
##   Odds ratio     1       0.88           0.64       0.46           
##   lower 95% CI           0.41           0.27       0.17           
##   upper 95% CI           1.9            1.47       1.15           
##                 pc1.q
## serostat         High SEP
##   hiv negative   541     
##   hiv  positive  7       
##                          
##   Odds ratio     0.41    
##   lower 95% CI   0.14    
##   upper 95% CI   1.04    
## 
## Chi-squared = 6.6 , 4 d.f., P value = 0.159 
## Fisher's exact test (2-sided) P value = 0.166

#--- Is there an association between Education & HIV Prevalence?
tz %$% tabpct(serostat, educat, percent = "col", graph = F)

## 
## Column percent 
##                educat
## serostat         no education, preschool       %  primary       %
##   hiv negative                       460  (96.2)     1743  (97.8)
##   hiv  positive                       18   (3.8)       39   (2.2)
##   Total                              478   (100)     1782   (100)
##                educat
## serostat         secondary       %
##   hiv negative         500  (99.6)
##   hiv  positive          2   (0.4)
##   Total                502   (100)

tz %$% cc(serostat, educat, graph = F)

## 
##                 educat
## serostat         no education, preschool primary secondary
##   hiv negative   460                     1743    500      
##   hiv  positive  18                      39      2        
##                                                           
##   Odds ratio     1                       0.57    0.1      
##   lower 95% CI                           0.32    0.01     
##   upper 95% CI                           1.07    0.43     
## 
## Chi-squared = 13.3 , 2 d.f., P value = 0.001 
## Fisher's exact test (2-sided) P value = 0.001