longitudinal_analysis_printout.Rmd

---
title: "Longitudinal Analysis"
author: 
    - "Francesco Beghini, Jackson Pullman, Marcus Alexander, Shivkumar Vishnempet Shridhar,"
    - "Drew Prinster, Adarsh Singh, Rigoberto Matute Juárez,"
    - "Edoardo M. Airoldi, Ilana L. Brito, Nicholas A. Christakis"
output: pdf_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE, warning = FALSE)
library(lmerTest)
```


```{r, echo=FALSE}
sharing_rate_only_followup <- readRDS('sharing_rate_only_followup.Rds')
sharing_rate_only_followup$village_code_ego <- as.factor(sharing_rate_only_followup$village_code_ego)
sharing_rate_only_followup$T2 <- scale(sharing_rate_only_followup$T2)
sharing_rate_only_followup$T1 <- scale(sharing_rate_only_followup$T1)
sharing_rate_only_followup$maha_dist <- scale(sharing_rate_only_followup$maha_dist)
xcov <- readRDS('xcov.RDS')
```

<!--
```{r, echo=FALSE}
lmer_zero_model <- lmer(T2 ~ T1 + relationship + (1|village_code_ego) + (1|ego), 
                          sharing_rate_only_followup)
summary(lmer_zero_model)
```
\newpage
-->
# First regression model
## Model the strain-sharing rate at timepoint 2 using the Mahalanobis distance as covariate
```{r}
lmer_first_model <- lmer(T2 ~ T1 + relationship + maha_dist + (1|village_code_ego) + (1|ego), 
                          sharing_rate_only_followup)
summary(lmer_first_model)
```
\newpage
## Model the strain-sharing rate at timepoint 2 using the separate sociodemographic variables
```{r, echo=FALSE}
lmer_first_model_expanded  <- lmer(T2 ~ T1 + relationship + 
                                      gender.mm + gender.ff + 
                                      indigenous.both + indigenous.one + 
                                      age_difference_abs + average_age  + 
                                      DDS_average + DDS_difference_abs + 
                                      education_difference_abs + average_education + 
                                      religion_same + building_same +
                                      wealth_difference_abs + average_wealth +
                                      average_bristol + different_bristol +
                                      watersource_same + same_usage_painkillers +
                                      same_usage_antibiotics +
                                      same_usage_antidiarrheal +
                                      same_usage_antiparasite +
                                      same_usage_vitamins +
                                      same_usage_zinc +
                                      same_usage_antifungal  + (1|village_code_ego) + (1|ego), data = sharing_rate_only_followup)

summary(lmer_first_model_expanded)
```
\newpage
# Second regression model
# Model each species sharing status at timepoint 2 using the Mahalanobis distance as covariate
```{r}
lmer_species_rand <- glmer(data = xcov, T2_t ~ T1_t + relationship + maha_dist + (1|species) + (1|village_code_ego) + (1|ego), family = binomial, nAGQ=0)
summary(lmer_species_rand)
```