Permalink
Cannot retrieve contributors at this time
Name already in use
A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?
Master-Thesis-Analyses/AntepartumDepression_ASQ_Total.R
Go to fileThis commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
2320 lines (2114 sloc)
105 KB
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| ### Results of prenatal depression and ASQ total | |
| #30.06.2021 | |
| ### Data preparation #### | |
| #setwd("C:/Users/User/Desktop/Internship/RScripts/Cortisol/Master_Thesis") | |
| setwd("C:/Users/alici/Desktop/Git_Folder/ITU_cortisol_analyses/Master_Thesis") | |
| ##libraries | |
| library(dplyr) | |
| library(gtsummary) | |
| library(gdata) | |
| library(VGAM) | |
| library(rcompanion) | |
| library(tidyverse) | |
| library(broom) | |
| library(APAstyler) | |
| ## 1. cortisol data #### | |
| load("Rdata/ITU_combined_cortisol_dates_times_wide_format.Rdata") | |
| comments_to_be_excluded <- c(88,6,12,3,15,9999,7,16,45,5,34,17,57,67,99) | |
| wide_cort <- wide_cort[!(wide_cort$notes %in% comments_to_be_excluded),] | |
| cort_IDs <- unique(wide_cort$participantID) | |
| ## 2. maternal well-being during pregnancy #### | |
| load("Rdata/processed_wellbeingduringpreg_completevars.Rdata") | |
| #for now only sleep, anxiety and CESD are included | |
| cols <- c("1", "2",grep("gestage", names(q_data_complete)), | |
| grep("GWdiff_cort_qclosest", names(q_data_complete)), | |
| grep("Cesd", names(q_data_complete)), | |
| #grep("puqe", names(q_data_complete)), | |
| grep("PSQI", names(q_data_complete)), | |
| #grep("ESS", names(q_data_complete)), | |
| grep("BAI",names(q_data_complete))) | |
| q_data_sub <- q_data_complete[,c(as.numeric(cols))] | |
| q_data_sub <- subset(q_data_sub, abs(GWdiff_cort_qclosest_to_cortGW) < 2) | |
| ## Depressive Symptoms Severity | |
| q_data_sub$Depressive_Symptom_Severity <- NA | |
| q_data_sub$Depressive_Symptom_Severity_num <- NA | |
| for(i in 1:nrow(q_data_sub)){ | |
| m <- q_data_sub$Cesd_qclosest_to_cortGW_pregMean[[i]] | |
| if(!is.na(m)){ | |
| if(m<16){ | |
| q_data_sub$Depressive_Symptom_Severity[[i]] <- "Below_Clinical (CES-D < 16)" | |
| q_data_sub$Depressive_Symptom_Severity_num[[i]] <- 0 | |
| } | |
| if(m>=16){ | |
| q_data_sub$Depressive_Symptom_Severity[[i]] <- "Clinical (CES-D >= 16)" | |
| q_data_sub$Depressive_Symptom_Severity_num[[i]] <- 1 | |
| } | |
| } | |
| } | |
| q_data_sub$Depressive_Symptom_Severity <- factor(q_data_sub$Depressive_Symptom_Severity) | |
| #Anxiety Symptom Severity | |
| q_data_sub$Anxiety_Symptom_Severity <- NA | |
| q_data_sub$Anxiety_Symptom_Severity_num <- NA | |
| for(i in 1:nrow(q_data_sub)){ | |
| m <- q_data_sub$BAI_qclosest_to_cortGW_pregMean[[i]] | |
| if(!is.na(m)){ | |
| if(m>21){ | |
| q_data_sub$Anxiety_Symptom_Severity[[i]] <- "Moderate/Severe" | |
| q_data_sub$Anxiety_Symptom_Severity_num[[i]] <- 1 | |
| } | |
| if(m<=21){ | |
| q_data_sub$Anxiety_Symptom_Severity[[i]] <- "Low/Normal" | |
| q_data_sub$Anxiety_Symptom_Severity_num[[i]] <- 0 | |
| } | |
| } | |
| } | |
| q_data_sub$psych_distress <- rowSums(q_data_sub[,c("Depressive_Symptom_Severity_num", | |
| "Anxiety_Symptom_Severity_num")], | |
| na.rm=F) | |
| ### compute additive depression score | |
| q_data_sub$clinD <- NA | |
| for(i in 1:nrow(q_data_sub)){ | |
| cesd <- q_data_sub$Cesd_qclosest_to_cortGW[[i]] | |
| if(!is.na(cesd)){ | |
| if(cesd >= 16){ | |
| q_data_sub$clinD[[i]] <- 1 | |
| } | |
| if(cesd < 16){ | |
| q_data_sub$clinD[[i]] <- 0 | |
| } | |
| } | |
| } | |
| ##sum occasions of clinical depression together | |
| q_data_sub <- q_data_sub %>% | |
| group_by(participantID) %>% | |
| mutate(additive_clinD = sum(clinD, na.rm = T)) | |
| ## data selection | |
| q_data_subx <- q_data_sub[,c("participantID", | |
| "pregstage", | |
| "Cesd_qclosest_to_cortGW_pregMean", | |
| "Depressive_Symptom_Severity", | |
| "Depressive_Symptom_Severity_num", | |
| "BAI_qclosest_to_cortGW_pregMean", | |
| "PSQI_qclosest_to_cortGW_pregMean", | |
| "qclosest_based_clin_Cesd", | |
| "Cesd_qclosest_to_cortGW", | |
| "BAI_qclosest_to_cortGW", | |
| "PSQI_qclosest_to_cortGW", | |
| "additive_clinD")] | |
| ## Sample stratification | |
| ## those with multiple assessments | |
| at_least_two_assessments <- c() | |
| q_data_subx <- as.data.frame(q_data_subx) | |
| IDs <- unique(q_data_subx$participantID[!is.na(q_data_subx$Cesd_qclosest_to_cortGW)]) | |
| for(i in 1:length(IDs)){ | |
| ID <- IDs[[i]] | |
| #print(ID) | |
| trims <- unique(q_data_subx[q_data_subx$participantID == ID, c("pregstage")]) | |
| #print(length(trims)) | |
| if(length(trims) >= 2){ | |
| at_least_two_assessments <- append(at_least_two_assessments, ID) | |
| } | |
| } | |
| ### Data in wide format | |
| q_data_sub_final <- q_data_subx %>% | |
| tidyr::pivot_wider( | |
| id_cols = c(participantID, | |
| pregstage, | |
| Cesd_qclosest_to_cortGW_pregMean, | |
| Depressive_Symptom_Severity, | |
| Depressive_Symptom_Severity_num, | |
| BAI_qclosest_to_cortGW_pregMean, | |
| PSQI_qclosest_to_cortGW_pregMean, | |
| additive_clinD), | |
| names_from = c(pregstage), # Can accommodate more variables, if needed. | |
| values_from = c(8:11) | |
| ) | |
| # #only overall questionnaire score during pregnancy | |
| # q_data_sub_final <- q_data_sub[,c("participantID", | |
| # "Cesd_qclosest_to_cortGW_pregMean", | |
| # "BAI_qclosest_to_cortGW_pregMean", | |
| # "PSQI_qclosest_to_cortGW_pregMean", | |
| # #"WS_CESD_variation_mean", | |
| # "Depressive_Symptom_Severity", | |
| # "Depressive_Symptom_Severity_num", | |
| # "Anxiety_Symptom_Severity", | |
| # "psych_distress")] | |
| # q_data_sub_final <- subset(q_data_sub_final, !duplicated(q_data_sub_final)) | |
| # summary(factor(q_data_sub_final$Anxiety_Symptom_Severity)) #n=5, these also have clin depression | |
| # summary(factor(q_data_sub_final$psych_distress)) | |
| # summary(factor(q_data_sub_final$Depressive_Symptom_Severity)) | |
| # Qs transformations #### | |
| q_data_sub_final$Cesd_qclosest_to_cortGW_pregMean_cent <- c(scale(sqrt(q_data_sub_final$Cesd_qclosest_to_cortGW_pregMean), scale = TRUE)) | |
| q_data_sub_final$BAI_qclosest_to_cortGW_pregMean_cent <- c(scale(sqrt(q_data_sub_final$BAI_qclosest_to_cortGW_pregMean), scale = TRUE)) | |
| q_data_sub_final$PSQI_qclosest_to_cortGW_pregMean_cent <- c(scale(sqrt(q_data_sub_final$PSQI_qclosest_to_cortGW_pregMean), scale = TRUE)) | |
| #add categorical anxiety severity | |
| q_data_sub_final$Anxiety_Severity_greateroneSD_num <- ifelse(q_data_sub_final$BAI_qclosest_to_cortGW_pregMean_cent > 1, 1, 0) | |
| q_data_sub_final$Anxiety_Severity_greateroneSD <- factor(ifelse(q_data_sub_final$BAI_qclosest_to_cortGW_pregMean_cent > 1, "yes", "no")) | |
| q_data_sub_final$Anxiety_Severity_greateroneSD <- factor(q_data_sub_final$Anxiety_Severity_greateroneSD) | |
| #add psych distress score | |
| q_data_sub_final$psych_distress <- rowSums(q_data_sub_final[,c("Depressive_Symptom_Severity_num", | |
| "Anxiety_Severity_greateroneSD_num")], | |
| na.rm=F) | |
| q_data_sub_final$psych_distress <- factor(q_data_sub_final$psych_distress) | |
| q_data_sub_final$PSQI_severity <- NA | |
| for(i in 1:nrow(q_data_sub_final)){ | |
| psqi <- q_data_sub_final$PSQI_qclosest_to_cortGW_pregMean_cent[[i]] | |
| if(!is.na(psqi)){ | |
| q_data_sub_final$PSQI_severity[[i]] <- "mean" | |
| if(psqi <= -1){ | |
| q_data_sub_final$PSQI_severity[[i]] <- "-1SD" | |
| } | |
| if(psqi >= 1){ | |
| q_data_sub_final$PSQI_severity[[i]] <- "+1SD" | |
| } | |
| } | |
| } | |
| q_data_sub_final$PSQI_severity <- factor(q_data_sub_final$PSQI_severity) | |
| ## 3. maternal follow-up data #### | |
| followUp <- read.delim("Rdata/ITU_1to2YearsFollowup_MaternalandChildQuestionnaires.dat") | |
| names(followUp)[1] <- "participantID" | |
| relevant_followUp <- c("1", | |
| grep("CESD", names(followUp)), | |
| grep("BAI", names(followUp))) | |
| postpartum_followUp <- followUp[,c(as.numeric(relevant_followUp))] | |
| postpartum_followUp$ITU_1.7y_mother_CESD_sum_nomis <- as.numeric(gsub(",", ".", postpartum_followUp$ITU_1.7y_mother_CESD_sum_nomis)) | |
| postpartum_followUp$ITU_1.7y_mother_BAI_sum_no_missing <- as.numeric(gsub(",", ".", postpartum_followUp$ITU_1.7y_mother_BAI_sum_no_missing)) | |
| postpartum_df <- postpartum_followUp[,c("participantID", | |
| "ITU_1.7y_mother_CESD_sum_nomis", | |
| "ITU_1.7y_mother_BAI_sum_no_missing")] | |
| names(postpartum_df)[2] <- "postpartum_Cesd" | |
| names(postpartum_df)[3] <- "postpartum_BAI" | |
| ## Depressive Symptoms Severity | |
| postpartum_df$postpartum_Depressive_Symptom_Severity <- NA | |
| postpartum_df$postpartum_Depressive_Symptom_Severity_num <- NA | |
| for(i in 1:nrow(postpartum_df)){ | |
| m <- postpartum_df$postpartum_Cesd[[i]] | |
| if(!is.na(m)){ | |
| if(m<16){ | |
| postpartum_df$postpartum_Depressive_Symptom_Severity[[i]] <- "Non-Clinical (CES-D < 16)" | |
| postpartum_df$postpartum_Depressive_Symptom_Severity_num[[i]] <- 0 | |
| } | |
| if(m>=16){ | |
| postpartum_df$postpartum_Depressive_Symptom_Severity[[i]] <- "Clinical (CES-D >= 16)" | |
| postpartum_df$postpartum_Depressive_Symptom_Severity_num[[i]] <- 1 | |
| } | |
| } | |
| } | |
| postpartum_df$postpartum_Depressive_Symptom_Severity <- factor(postpartum_df$postpartum_Depressive_Symptom_Severity) | |
| #view(postpartum_df[,c("postpartum_Cesd","postpartum_Depressive_Symptom_Severity_num")]) | |
| #Anxiety Symptom Severity | |
| postpartum_df$postpartum_Anxiety_Symptom_Severity <- NA | |
| postpartum_df$postpartum_Anxiety_Symptom_Severity_num <- NA | |
| for(i in 1:nrow(postpartum_df)){ | |
| m <- postpartum_df$postpartum_BAI[[i]] | |
| if(!is.na(m)){ | |
| if(m>21){ | |
| postpartum_df$postpartum_Anxiety_Symptom_Severity[[i]] <- "Moderate/Severe" | |
| postpartum_df$postpartum_Anxiety_Symptom_Severity_num[[i]] <- 1 | |
| } | |
| if(m<=21){ | |
| postpartum_df$postpartum_Anxiety_Symptom_Severity[[i]] <- "Low/Normal" | |
| postpartum_df$postpartum_Anxiety_Symptom_Severity_num[[i]] <- 0 | |
| } | |
| } | |
| } | |
| #summary(factor(postpartum_df$postpartum_Anxiety_Symptom_Severity)) n = 5 | |
| #summary(factor(postpartum_df$postpartum_Depressive_Symptom_Severity)) n = 102 | |
| ### transformations | |
| postpartum_df$postpartum_BAI_cent <- c(scale(sqrt(postpartum_df$postpartum_BAI), scale = TRUE)) | |
| postpartum_df$postpartum_Cesd_cent <- c(scale(sqrt(postpartum_df$postpartum_Cesd), scale = TRUE)) | |
| ### anxiety severity by SD above the sample mean | |
| postpartum_df$postpartum_Anxiety_Severity_greateroneSD_num <- ifelse(postpartum_df$postpartum_BAI_cent > 1, 1, 0) | |
| postpartum_df$postpartum_Anxiety_Severity_greateroneSD <- factor(ifelse(postpartum_df$postpartum_BAI_cent > 1, "yes", "no")) | |
| postpartum_df$postpartum_psych_distress <- rowSums(postpartum_df[,c("postpartum_Depressive_Symptom_Severity_num", | |
| "postpartum_Anxiety_Severity_greateroneSD_num")], | |
| na.rm=F) | |
| postpartum_df$postpartum_psych_distress <- factor(postpartum_df$postpartum_psych_distress) | |
| postpartum_df_final <- postpartum_df[,c("participantID", | |
| "postpartum_BAI_cent", | |
| "postpartum_Cesd_cent", | |
| "postpartum_Depressive_Symptom_Severity", | |
| "postpartum_Anxiety_Severity_greateroneSD", | |
| "postpartum_psych_distress", | |
| "postpartum_Depressive_Symptom_Severity_num")] | |
| # 3.1 Maternal Education #### | |
| maternal_edu <- read.delim("Rdata/ITU maternal education.dat") | |
| names(maternal_edu)[1] <- "participantID" | |
| names(maternal_edu)[3] <- "Maternal_Education" | |
| maternal_edu <- maternal_edu[,c("participantID", | |
| "Maternal_Education")] | |
| library(naniar) | |
| maternal_edu <- maternal_edu %>% replace_with_na(replace = list(Maternal_Education = -9)) | |
| maternal_edu$Maternal_Education <- factor(maternal_edu$Maternal_Education, | |
| levels = c(1,2,3), | |
| labels = c("primary", "applied university", "university")) | |
| ## 4. Register data during pregnancy #### | |
| load("Rdata/processed_register_data.Rdata") | |
| register_data$Maternal_Smoking_During_Pregnancy <- factor(register_data$Maternal_Smoking_During_Pregnancy, | |
| levels = c("no", "quit_T1", "yes"), | |
| labels = c("no", "no", "yes")) | |
| register_data$Maternal_Hypertensive_Disorders_anyVSnone <- factor(register_data$Maternal_Hypertensive_Disorders_anyVSnone, | |
| levels = c(-999,0,1), | |
| labels = c("no", "no", "yes")) | |
| register_data$Maternal_Diabetes_Disorders_anyVSnone <- factor(register_data$Maternal_Diabetes_Disorders_anyVSnone, | |
| levels = c(-999,0,1), | |
| labels = c("no", "no", "yes")) | |
| register_data$Maternal_Body_Mass_Index_in_Early_Pregnancy_cent <- c(scale(register_data$Maternal_Body_Mass_Index_in_Early_Pregnancy, scale = F)) | |
| register_data$Weight_Gain_cent <- c(scale(register_data$Weight_Gain, scale = F)) | |
| register_data$Gestational_Age_Weeks_cent <- c(scale(register_data$Gestational_Age_Weeks, scale = F)) | |
| register_data$Child_Birth_Weight_cent <- c(scale(register_data$Child_Birth_Weight, scale = F)) | |
| register_data$Maternal_Age_Years_cent <- c(scale(register_data$Maternal_Age_Years, scale = F)) | |
| regis_final <- register_data[,c("participantID", | |
| "caseVScontrol", | |
| "Parity", | |
| "Maternal_Smoking_During_Pregnancy", | |
| "Maternal_Corticosteroid_Treatment_during_Pregnancy", | |
| "Maternal_Body_Mass_Index_in_Early_Pregnancy_cent", | |
| "Child_Sex" , | |
| "Gestational_Age_Weeks_cent", | |
| "Child_Birth_Weight_cent", | |
| "Weight_Gain_cent", | |
| "Maternal_Hypertensive_Disorders_anyVSnone", | |
| "Maternal_Diabetes_Disorders_anyVSnone", | |
| "Maternal_Age_Years_cent" | |
| )] | |
| ## 4.1 medication data #### | |
| medication <- read.delim("Rdata/ITU_psychotrophicmedication05July21_Maternal_CurrentPregnancy.dat") | |
| names(medication)[1] <- "participantID" | |
| medication$ITUbroadpsychiatricmedication_18_KELA <- factor(medication$ITUbroadpsychiatricmedication_18_KELA, | |
| levels = c(0,1), | |
| labels = c("no", "yes")) | |
| ## 5. ASQ data #### | |
| ASQ <- read.csv2("Rdata/ASQ_dataset_ITU_01122020r_shorter.csv") | |
| names(ASQ)[1] <- "participantID" | |
| #extract only participants who have cortisol data and finalagerange scores | |
| #ASQ_final <- ASQ[ASQ$participantID %in% cort_IDs,c(1,3, 18:23)] | |
| ASQ_final <- ASQ[ASQ$participantID %in% cort_IDs,c(1,3, 18:23)] | |
| ASQ_final_samplesizes <- left_join(ASQ_final, wide_cort) | |
| ASQ_final$Child_ASQ_grossmotor_development_infancy_sum_finalagerange <- | |
| as.numeric(ASQ_final$Child_ASQ_grossmotor_development_infancy_sum_finalagerange) | |
| #normalized rank scores | |
| for(col in 3:ncol(ASQ_final)){ | |
| column <- ASQ_final[col] | |
| name <- paste0(names(ASQ_final[col]), "_norm") | |
| ASQ_final[name] <- blom(column, method = "rankit") | |
| } | |
| #dichotomize normalized rank scores | |
| for(col in 9:ncol(ASQ_final)){ | |
| column <- ASQ_final[col] | |
| old_name <- names(ASQ_final[col]) | |
| new_name <- paste0(names(ASQ_final[col]), "_dichom") | |
| ASQ_final[new_name] <- NA | |
| for(i in 1:nrow(ASQ_final)){ | |
| ASQ_score <- ASQ_final[[i, c(old_name)]] | |
| if(!is.na(ASQ_score)){ | |
| if(ASQ_score <= -1){ | |
| ASQ_final[[i,c(new_name)]] <- 1 | |
| } | |
| if(ASQ_score > -1){ | |
| ASQ_final[[i,c(new_name)]] <- 0 | |
| } | |
| } | |
| } | |
| } | |
| ASQ_final$ChildAge_ASQ_months_allchildren_cent <- | |
| c(scale(ASQ_final$ChildAge_ASQ_months_allchildren, scale = F)) | |
| #select subset of relevant variables | |
| ASQ_final <- ASQ_final[,c("participantID", | |
| "ChildAge_ASQ_months_allchildren_cent", | |
| "ChildAge_ASQ_months_allchildren", | |
| "Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom", | |
| "Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange")] | |
| ## 6. On Pregnancy Averages: merge all data #### | |
| #use all the ASQ data | |
| ASQ_df1 <- left_join(ASQ_final, q_data_sub_final) | |
| ASQ_df2 <- left_join(ASQ_df1, maternal_edu) | |
| ASQ_df3 <- left_join(ASQ_df2, postpartum_df_final) | |
| ASQ_df_final <- left_join(ASQ_df3, regis_final) | |
| ASQ_df_final <- left_join(ASQ_df_final, | |
| medication[,c("participantID", | |
| "ITUbroadpsychiatricmedication_18_KELA", | |
| "antidepressants_18_KELA")], | |
| copy=TRUE) | |
| #exclude violattions against cortisol collection protocol | |
| ASQ_df_final <- ASQ_df_final[!(ASQ_df_final$Maternal_Corticosteroid_Treatment_during_Pregnancy == "yes"),] | |
| # length(unique(ASQ_df1$participantID[!is.na(ASQ_df_final$Cesd_qclosest_to_cortGW_pregMean)])) | |
| # length(unique(ASQ_df1$participantID[!is.na(ASQ_df_final$Cesd_qclosest_to_cortGW_I)])) | |
| # length(unique(ASQ_df1$participantID[!is.na(ASQ_df_final$Cesd_qclosest_to_cortGW_II)])) | |
| # length(unique(ASQ_df1$participantID[!is.na(ASQ_df_final$Cesd_qclosest_to_cortGW_III)])) | |
| ##additive pre/post effects | |
| ASQ_df_final$Pre.Post_clinD <- NA | |
| for(i in 1:nrow(ASQ_df_final)){ | |
| pre <- ASQ_df_final$Depressive_Symptom_Severity_num[[i]] | |
| post <- ASQ_df_final$postpartum_Depressive_Symptom_Severity_num[[i]] | |
| if(!is.na(pre) & !is.na(post) & pre == 0 & post == 0){ | |
| ASQ_df_final$Pre.Post_clinD[[i]] <- "never" | |
| } | |
| if(sum(pre, post, na.rm=T) == 1 & !is.na(pre) & pre == 1){ | |
| ASQ_df_final$Pre.Post_clinD[[i]] <- "pre_only" | |
| } | |
| if(sum(pre, post, na.rm=T) == 1 & !is.na(post) & post == 1){ | |
| ASQ_df_final$Pre.Post_clinD[[i]] <- "post_only" | |
| } | |
| if(sum(pre, post, na.rm=T) == 2){ | |
| ASQ_df_final$Pre.Post_clinD[[i]] <- "pre_post" | |
| } | |
| } | |
| ASQ_df_final$Pre.Post_clinD <- factor(ASQ_df_final$Pre.Post_clinD) | |
| #view(ASQ_df_final[,c("Depressive_Symptom_Severity_num", "postpartum_Depressive_Symptom_Severity_num", "Pre.Post_clinD")]) | |
| ## 7. Apply Exclusion Criteria #### | |
| ASQ_df_final <- subset(ASQ_df_final, !(Maternal_Corticosteroid_Treatment_during_Pregnancy == "yes")) | |
| #final n = 528 | |
| # ASQ_numbers <- ASQ_df_final[, c("Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom", | |
| # "Depressive_Symptom_Severity")] | |
| # | |
| # tbl_summary(data = ASQ_numbers, | |
| # by = Depressive_Symptom_Severity) | |
| #tidy up | |
| rm(list= ls()[!(ls() %in% c("ASQ_df_final", "at_least_two_assessments"))]) | |
| ASQ_df_final_sub <- ASQ_df_final[!(ASQ_df_final$ITUbroadpsychiatricmedication_18_KELA == "yes"),] | |
| ## nr of CESD assessments per mother #### | |
| # ASQ_df_final$nr_cesds <- NA | |
| # for(i in 1:nrow(ASQ_df_final)){ | |
| # cesdI <- ASQ_df_final$Cesd_qclosest_to_cortGW_I[[i]] | |
| # cesdII <- ASQ_df_final$Cesd_qclosest_to_cortGW_II[[i]] | |
| # cesdIII <- ASQ_df_final$Cesd_qclosest_to_cortGW_III[[i]] | |
| # | |
| # I <- ifelse(!is.na(cesdI), 1,0) | |
| # II <- ifelse(!is.na(cesdII), 1,0) | |
| # III <- ifelse(!is.na(cesdIII), 1,0) | |
| # | |
| # ASQ_df_final$nr_cesds [[i]] <- I + II + III | |
| # } | |
| # save(ASQ_df_final, file = "Rdata/nr_cesd.Rdata") | |
| ################################################################################ | |
| ## Set up results table #### | |
| ASQ_total_results <- setNames(data.frame(matrix(ncol = 13, nrow = 8)), | |
| c("Maternal depressive symptoms", | |
| "B", | |
| "SE", | |
| "LL", | |
| "UL", | |
| "z", | |
| "p", | |
| "OR", | |
| "SE", | |
| "LL", | |
| "UL", | |
| "z", | |
| "p")) | |
| ASQ_total_results[,1] <- c("Pregnancy mean CES-D score (n = ", | |
| "Model 1", | |
| "Model 2", | |
| "Model 3", | |
| "Pregnancy mean CES-D score >= 16", | |
| "Model 1", | |
| "Model 2", | |
| "Model 3") | |
| ## Set up functions to report statistics in APA #### | |
| tobit_apa <- function(m,x){ #m=tobit model, x=predictor as displayed in model summary table | |
| #document results | |
| estimate <- format(round(coef(summary(m))[x,1],2)) | |
| SE <- format(round(coef(summary(m))[x,2],2)) | |
| CI <- c(confint(m,x)) | |
| LL <- format(round(CI[1],2)) | |
| UL <- format(round(CI[2],2)) | |
| z <- format(round(coef(summary(m))[x,3],2)) | |
| p <- snip(as.numeric(format(round(coef(summary(m))[x,4],3))), lead=1) | |
| output <- list(estimate,SE,LL,UL,z,p) | |
| names(output) <- c("estimate","SE","LL","UL","z","p") | |
| return(output) #output = list object of any parameters that may be interesting to report | |
| } | |
| logit_apa <- function(m,x){ #m=lobit model, x=predictor as displayed in model summary table | |
| #document results | |
| estimate <- format(round(exp(coef(summary(m))[x,1]),2)) | |
| SE <- format(round(exp(coef(summary(m))[x,2]),2)) | |
| CI <- exp(c(confint(m,x))) | |
| LL <- format(round(CI[1],2)) | |
| UL <- format(round(CI[2],2)) | |
| z <- format(round(coef(summary(m))[x,3],2)) | |
| p <- snip(as.numeric(format(round(coef(summary(m))[x,4],3))), lead = 1) | |
| output <- list(estimate,SE,LL,UL,z,p) | |
| names(output) <- c("estimate","SE","LL","UL","z","p") | |
| return(output) #output = list object of any parameters (of OR) that may be interesting to report | |
| } | |
| ################################################################################ | |
| #### Tobit on Pregnancy Averages #### | |
| ## CES-D #### | |
| M.CESD <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Cesd_qclosest_to_cortGW_pregMean_cent, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final, | |
| na.action = "na.exclude") | |
| summary(M.CESD) | |
| res <- tobit_apa(M.CESD, "Cesd_qclosest_to_cortGW_pregMean_cent") | |
| ASQ_total_results[2,2] <- res$estimate | |
| ASQ_total_results[2,3] <- res$SE | |
| ASQ_total_results[2,4] <- res$LL | |
| ASQ_total_results[2,5] <- res$UL | |
| ASQ_total_results[2,6] <- res$z | |
| ASQ_total_results[2,7] <- res$p | |
| M.PSQI <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| PSQI_qclosest_to_cortGW_pregMean_cent, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final, | |
| na.action = "na.exclude") | |
| summary(M.PSQI) | |
| #document results | |
| Tobit_res.PSQI <- tobit_apa(M.PSQI, "PSQI_qclosest_to_cortGW_pregMean_cent") | |
| M.BAI <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| BAI_qclosest_to_cortGW_pregMean_cent, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final, | |
| na.action = "na.exclude") | |
| summary(M.BAI) | |
| #document results | |
| Tobit_res.BAI <- tobit_apa(M.BAI, "BAI_qclosest_to_cortGW_pregMean_cent") | |
| ##test assumptions | |
| ASQ_df_final$yhat <- fitted(M.CESD)[,1] | |
| ASQ_df_final$rr <- resid(M.CESD, type = "response") | |
| ASQ_df_final$rp <- resid(M.CESD, type = "pearson")[,1] | |
| #par(mfcol = c(2, 3)) | |
| # with(ASQ_df_final, { | |
| # plot(yhat, rr, main = "Fitted vs Residuals") | |
| # qqnorm(rr) | |
| # plot(yhat, rp, main = "Fitted vs Pearson Residuals") | |
| # qqnorm(rp) | |
| # plot(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange, | |
| # rp, main = "Actual vs Pearson Residuals") | |
| # plot(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange, | |
| # yhat, main = "Actual vs Fitted") | |
| # }) | |
| # summary(M.CESD) | |
| #Sex-specific effects | |
| M.CESDINT_ChS <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Cesd_qclosest_to_cortGW_pregMean_cent + | |
| Cesd_qclosest_to_cortGW_pregMean_cent:Child_Sex, | |
| tobit(Upper = 300), data = ASQ_df_final) | |
| summary(M.CESDINT_ChS) | |
| #document results | |
| Tobit_res.CESDxChS <- tobit_apa(M.CESDINT_ChS, "Child_Sexgirl:Cesd_qclosest_to_cortGW_pregMean_cent") | |
| #caseVScontrol | |
| M.CESDINT_CaCo <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| caseVScontrol + | |
| Cesd_qclosest_to_cortGW_pregMean_cent + | |
| Cesd_qclosest_to_cortGW_pregMean_cent:caseVScontrol, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final) | |
| summary(M.CESDINT_CaCo) | |
| #document results | |
| Tobit_res.CESDxCaCo <- tobit_apa(M.CESDINT_CaCo, "caseVScontrolcontrol:Cesd_qclosest_to_cortGW_pregMean_cent") | |
| ### adding prenatal covariates | |
| M.CESD.cov1 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Cesd_qclosest_to_cortGW_pregMean_cent + | |
| #BAI_qclosest_to_cortGW_pregMean_cent + | |
| #PSQI_qclosest_to_cortGW_pregMean_cent + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent, | |
| tobit(Upper = 300), | |
| na.action = "na.exclude", | |
| data = ASQ_df_final) | |
| summary(M.CESD.cov1) | |
| #document results | |
| res <- tobit_apa(M.CESD.cov1, "Cesd_qclosest_to_cortGW_pregMean_cent") | |
| ASQ_total_results[3,2] <- res$estimate | |
| ASQ_total_results[3,3] <- res$SE | |
| ASQ_total_results[3,4] <- res$LL | |
| ASQ_total_results[3,5] <- res$UL | |
| ASQ_total_results[3,6] <- res$z | |
| ASQ_total_results[3,7] <- res$p | |
| #total variance explained by tobit model | |
| ASQ_df_final$yhat <- fitted(M.CESD.cov1)[,1] | |
| r <- cor(ASQ_df_final$yhat, | |
| ASQ_df_final$Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange, | |
| use = "pairwise.complete.obs") | |
| Tobit_var.explained.prenat <- format(round(r^2,2)) | |
| ## adding the postnatal factors | |
| M.CESD.cov2 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Cesd_qclosest_to_cortGW_pregMean_cent + | |
| #BAI_qclosest_to_cortGW_pregMean_cent + | |
| #PSQI_qclosest_to_cortGW_pregMean_cent + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent + | |
| postpartum_Cesd_cent + | |
| postpartum_BAI_cent, | |
| tobit(Upper = 300), | |
| na.action = "na.exclude", | |
| data = ASQ_df_final) | |
| summary(M.CESD.cov2) | |
| #document results | |
| res <- tobit_apa(M.CESD.cov2, "Cesd_qclosest_to_cortGW_pregMean_cent") | |
| ASQ_total_results[4,2] <- res$estimate | |
| ASQ_total_results[4,3] <- res$SE | |
| ASQ_total_results[4,4] <- res$LL | |
| ASQ_total_results[4,5] <- res$UL | |
| ASQ_total_results[4,6] <- res$z | |
| ASQ_total_results[4,7] <- res$p | |
| ## also report sig. covariates | |
| # COV.tobit_ChS <- tobit_apa(M.CESD.cov2, "Child_Sexgirl") | |
| # COV.tobit_Diab <- tobit_apa(M.CESD.cov2, "Maternal_Diabetes_Disorders_anyVSnoneyes") | |
| # COV.tobit_birthW <- tobit_apa(M.CESD.cov2, "Child_Birth_Weight_cent") | |
| # COV.tobit_postD <- tobit_apa(M.CESD.cov2, "postpartum_Cesd_cent") | |
| #total variance explained by tobit model | |
| ASQ_df_final$yhat <- fitted(M.CESD.cov2)[,1] | |
| r <- cor(ASQ_df_final$yhat, | |
| ASQ_df_final$Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange, | |
| use = "pairwise.complete.obs") | |
| Tobit_var.explained_post <- r^2 | |
| ## categorical depression #### | |
| M.CESD_cat <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Depressive_Symptom_Severity, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final) | |
| summary(M.CESD_cat) | |
| #document results | |
| res <- tobit_apa(M.CESD_cat, "Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| ASQ_total_results[6,2] <- res$estimate | |
| ASQ_total_results[6,3] <- res$SE | |
| ASQ_total_results[6,4] <- res$LL | |
| ASQ_total_results[6,5] <- res$UL | |
| ASQ_total_results[6,6] <- res$z | |
| ASQ_total_results[6,7] <- res$p | |
| ##INT childsex | |
| M.CESD_cat.INT1 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Depressive_Symptom_Severity + | |
| Depressive_Symptom_Severity:Child_Sex, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final) | |
| summary(M.CESD_cat.INT1) | |
| #document results | |
| Tobit_res.CESDxChS_cat <- tobit_apa(M.CESD_cat.INT1, "Child_Sexgirl:Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| #INT caseVScontrol | |
| M.CESD_cat.INT2 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| caseVScontrol + | |
| Depressive_Symptom_Severity + | |
| Depressive_Symptom_Severity:caseVScontrol, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final) | |
| summary(M.CESD_cat.INT2) | |
| #document | |
| Tobit_res.CESDxCaCo_cat <- tobit_apa(M.CESD_cat.INT2, "caseVScontrolcontrol:Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| ## incl. prenatal covariates | |
| M.CESD_cat.cov1 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Depressive_Symptom_Severity + | |
| #BAI_qclosest_to_cortGW_pregMean_cent + | |
| #PSQI_qclosest_to_cortGW_pregMean_cent + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final) | |
| summary(M.CESD_cat.cov1) | |
| #document results | |
| res <- tobit_apa(M.CESD_cat.cov1, "Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| ASQ_total_results[7,2] <- res$estimate | |
| ASQ_total_results[7,3] <- res$SE | |
| ASQ_total_results[7,4] <- res$LL | |
| ASQ_total_results[7,5] <- res$UL | |
| ASQ_total_results[7,6] <- res$z | |
| ASQ_total_results[7,7] <- res$p | |
| ## incl. postpartum covariates | |
| M.CESD_cat.cov2 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Depressive_Symptom_Severity + | |
| #BAI_qclosest_to_cortGW_pregMean_cent + | |
| #PSQI_qclosest_to_cortGW_pregMean_cent + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent + | |
| postpartum_BAI_cent + | |
| postpartum_Cesd_cent, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final) | |
| summary(M.CESD_cat.cov2) | |
| ##document results | |
| res <- tobit_apa(M.CESD_cat.cov2, "Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| ASQ_total_results[8,2] <- res$estimate | |
| ASQ_total_results[8,3] <- res$SE | |
| ASQ_total_results[8,4] <- res$LL | |
| ASQ_total_results[8,5] <- res$UL | |
| ASQ_total_results[8,6] <- res$z | |
| ASQ_total_results[8,7] <- res$p | |
| ## additive pre-post clinical depression #### | |
| M.CESD_add <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final, | |
| na.action = "na.exclude") | |
| summary(M.CESD_add) | |
| ##document results | |
| tobit_res.addD_pre <- tobit_apa(M.CESD_add, "Pre.Post_clinDpre_only") | |
| tobit_res.addD_post <- tobit_apa(M.CESD_add, "Pre.Post_clinDpost_only") | |
| tobit_res.addD_preANDpost <- tobit_apa(M.CESD_add, "Pre.Post_clinDpre_post") | |
| M.CESD_add.vsPOST <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| relevel(Pre.Post_clinD, ref = "post_only"), | |
| tobit(Upper = 300), | |
| data = ASQ_df_final, | |
| na.action = "na.exclude") | |
| rownames(coef(summary(M.CESD_add.vsPOST))) | |
| #results relative to post-only | |
| tobit_res.addD_pre_post.VS.post_only <- tobit_apa(M.CESD_add.vsPOST, "relevel(Pre.Post_clinD, ref = \"post_only\")pre_post") | |
| ##test assumptions | |
| ASQ_df_final$yhat <- fitted(M.CESD_add)[,1] | |
| ASQ_df_final$rr <- resid(M.CESD_add, type = "response") | |
| ASQ_df_final$rp <- resid(M.CESD_add, type = "pearson")[,1] | |
| # par(mfcol = c(2, 3)) | |
| # | |
| # with(ASQ_df_final, { | |
| # plot(yhat, rr, main = "Fitted vs Residuals") | |
| # qqnorm(rr) | |
| # plot(yhat, rp, main = "Fitted vs Pearson Residuals") | |
| # qqnorm(rp) | |
| # plot(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange, | |
| # rp, main = "Actual vs Pearson Residuals") | |
| # plot(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange, | |
| # yhat, main = "Actual vs Fitted") | |
| # }) | |
| summary(M.CESD) | |
| #Sex-specific effects | |
| M.CESDINT_ChS_add1 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final) | |
| M.CESDINT_ChS_add2 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD + | |
| Pre.Post_clinD:Child_Sex, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final) | |
| p <- pchisq(2 * (logLik(M.CESDINT_ChS_add1) - logLik(M.CESDINT_ChS_add2)), | |
| df = 2, | |
| lower.tail = FALSE) | |
| Tobit_res.CESDxChS_add <- list(logLik(M.CESDINT_ChS_add1),logLik(M.CESDINT_ChS_add2), p) | |
| #caseVScontrol | |
| #without interaction | |
| M.CESDINT_CaCo_add1 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| caseVScontrol + | |
| Pre.Post_clinD, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final) | |
| summary(M.CESDINT_CaCo_add1) | |
| #with interaction | |
| M.CESDINT_CaCo_add2 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| caseVScontrol + | |
| Pre.Post_clinD + | |
| Pre.Post_clinD:caseVScontrol, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final) | |
| summary(M.CESDINT_CaCo_add2) | |
| ##document results | |
| p <- pchisq(2 * (logLik(M.CESDINT_CaCo_add1) - logLik(M.CESDINT_CaCo_add2)), | |
| df = 2, | |
| lower.tail = FALSE) | |
| Tobit_res.CESDxCaCo_add <- list(logLik(M.CESDINT_CaCo_add1),logLik(M.CESDINT_CaCo_add2), p) | |
| ### adding prenatal covariates | |
| M.CESD.cov1_add <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD + | |
| #BAI_qclosest_to_cortGW_pregMean_cent + | |
| #PSQI_qclosest_to_cortGW_pregMean_cent + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final) | |
| summary(M.CESD.cov1_add) | |
| ## adding the postnatal factors | |
| M.CESD.cov2_add <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD + | |
| #BAI_qclosest_to_cortGW_pregMean_cent + | |
| #PSQI_qclosest_to_cortGW_pregMean_cent + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent + | |
| postpartum_BAI_cent, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final) | |
| summary(M.CESD.cov2_add) | |
| ################################################################################ | |
| ###### Logistic Regression on Neurodevelopmental delay #### | |
| # assumptions #### | |
| # new_ASQ.total <- na.omit(ASQ_df_final[,-c(4, 7:18,22)]) #first exclude the pregnancy-stage specific predictors | |
| # # # Fit the logistic regression model | |
| # model <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| # ChildAge_ASQ_months_allchildren_cent + | |
| # Child_Sex + | |
| # Cesd_qclosest_to_cortGW_pregMean_cent + | |
| # BAI_qclosest_to_cortGW_pregMean_cent + | |
| # PSQI_qclosest_to_cortGW_pregMean_cent + | |
| # caseVScontrol + | |
| # Maternal_Age_Years_cent + | |
| # Maternal_Education + | |
| # Parity + | |
| # Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| # Weight_Gain_cent + | |
| # Maternal_Hypertensive_Disorders_anyVSnone + | |
| # Maternal_Diabetes_Disorders_anyVSnone + | |
| # Maternal_Smoking_During_Pregnancy + | |
| # Gestational_Age_Weeks_cent + | |
| # Child_Birth_Weight_cent + | |
| # postpartum_Cesd_cent + | |
| # postpartum_BAI_cent, | |
| # data = new_ASQ.total, | |
| # family = binomial, | |
| # na.action = "na.exclude") | |
| # | |
| # # Predict the probability (p) of diabete positivity | |
| # probabilities <- predict(model, type = "response") | |
| # # predicted.classes <- ifelse(probabilities > 0.3, "pos", "neg") | |
| # # head(predicted.classes) | |
| # # Select only numeric predictors | |
| # mydata <- new_ASQ.total %>% | |
| # dplyr::select_if(is.numeric) | |
| # #mydata <- mydata[,-c(5:10)] | |
| # predictors <- colnames(mydata) | |
| # # Bind the logit and tidying the data for plot | |
| # mydata <- mydata %>% | |
| # mutate(logit = log(probabilities/(1-probabilities))) %>% | |
| # gather(key = "predictors", value = "predictor.value", -logit) | |
| # | |
| # #create scatterplots | |
| # ggplot(mydata, aes(logit, predictor.value))+ | |
| # geom_point(size = 0.5, alpha = 0.5) + | |
| # geom_smooth(method = "loess") + | |
| # theme_bw() + | |
| # facet_wrap(~predictors, scales = "free_y") | |
| # | |
| # ## Cook's distance | |
| # plot(model, which = 4, id.n = 3) | |
| # | |
| # #influential cases | |
| # # Extract model results | |
| # #new_ASQ <- ASQ_df_final[,-c(10:14)] #first exclude the other outcome variables | |
| # model.data <- augment(model, data = new_ASQ.total) %>% mutate(index = 1:n()) | |
| # outliers <- model.data %>% top_n(3, .cooksd) #with with neurodevelopmental delay, high depression and anxiety (two controls, one case) | |
| # | |
| # #plot standardize residuals | |
| # ggplot(model.data, aes(index, .std.resid)) + | |
| # geom_point(aes(color = Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom), alpha = .5) + | |
| # theme_bw() | |
| # model.data %>% | |
| # filter(abs(.std.resid) > 3) #no influential cases | |
| # | |
| # #multicollinearity | |
| # car::vif(model) | |
| ## Continuous CES-D Models #### | |
| #ASQ and prenatal Cesd | |
| M1_total <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Cesd_qclosest_to_cortGW_pregMean_cent, | |
| data = ASQ_df_final, | |
| family = "binomial") | |
| summary(M1_total) | |
| res <- logit_apa(M1_total,"Cesd_qclosest_to_cortGW_pregMean_cent") | |
| ASQ_total_results[2,8] <- res$estimate | |
| ASQ_total_results[2,9] <- res$SE | |
| ASQ_total_results[2,10] <- res$LL | |
| ASQ_total_results[2,11] <- res$UL | |
| ASQ_total_results[2,12] <- res$z | |
| ASQ_total_results[2,13] <- res$p | |
| #ASQ and BAI | |
| M1_total_BAI <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| BAI_qclosest_to_cortGW_pregMean_cent, | |
| data = ASQ_df_final, | |
| family = "binomial") | |
| summary(M1_total_BAI) | |
| logit.res_BAI <- logit_apa(M1_total_BAI,"BAI_qclosest_to_cortGW_pregMean_cent") | |
| #ASQ and PSQI | |
| M1_total_PSQI <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| PSQI_qclosest_to_cortGW_pregMean_cent, | |
| data = ASQ_df_final, | |
| family = "binomial") | |
| summary(M1_total_PSQI) | |
| logit.res_PSQI <- logit_apa(M1_total_PSQI,"PSQI_qclosest_to_cortGW_pregMean_cent") | |
| #ASQ and prenatal Cesd*child sex | |
| M1_total.INT1 <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Cesd_qclosest_to_cortGW_pregMean_cent + | |
| Cesd_qclosest_to_cortGW_pregMean_cent:Child_Sex, | |
| data = ASQ_df_final, | |
| family = "binomial") | |
| summary(M1_total.INT1) | |
| logit.res_CESDxChS <- logit_apa(M1_total.INT1,"Child_Sexgirl:Cesd_qclosest_to_cortGW_pregMean_cent") | |
| #ASQ and prenatal Cesd*caseVScontrol | |
| M1_total.INT2 <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| caseVScontrol + | |
| Cesd_qclosest_to_cortGW_pregMean_cent + | |
| Cesd_qclosest_to_cortGW_pregMean_cent:caseVScontrol, | |
| data = ASQ_df_final, | |
| family = "binomial") | |
| summary(M1_total.INT2) | |
| logit.res_CESDxCaCo <- logit_apa(M1_total.INT2,"caseVScontrolcontrol:Cesd_qclosest_to_cortGW_pregMean_cent") | |
| #ASQ and prenatal Cesd + prenatal covariates | |
| M2_total.cov <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Cesd_qclosest_to_cortGW_pregMean_cent + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent, | |
| data = ASQ_df_final, | |
| family = "binomial") | |
| summary(M2_total.cov) | |
| #document | |
| res <- logit_apa(M2_total.cov,"Cesd_qclosest_to_cortGW_pregMean_cent") | |
| ASQ_total_results[3,8] <- res$estimate | |
| ASQ_total_results[3,9] <- res$SE | |
| ASQ_total_results[3,10] <- res$LL | |
| ASQ_total_results[3,11] <- res$UL | |
| ASQ_total_results[3,12] <- res$z | |
| ASQ_total_results[3,13] <- res$p | |
| #ASQ and prenatal Cesd + postnatal covariates | |
| M3_total.cov <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Cesd_qclosest_to_cortGW_pregMean_cent + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent + | |
| postpartum_BAI_cent + | |
| postpartum_Cesd_cent, | |
| data = ASQ_df_final, | |
| family = "binomial") | |
| summary(M3_total.cov) | |
| COV.logit_maternal_edu <- logit_apa(M3_total.cov, "Maternal_Educationuniversity") | |
| COV.logit_birthW <- logit_apa(M3_total.cov, "Child_Birth_Weight_cent") | |
| #document | |
| res <- logit_apa(M3_total.cov,"Cesd_qclosest_to_cortGW_pregMean_cent") | |
| ASQ_total_results[4,8] <- res$estimate | |
| ASQ_total_results[4,9] <- res$SE | |
| ASQ_total_results[4,10] <- res$LL | |
| ASQ_total_results[4,11] <- res$UL | |
| ASQ_total_results[4,12] <- res$z | |
| ASQ_total_results[4,13] <- res$p | |
| ## Categorical CES-D Models ##### | |
| # assumptions | |
| # new_ASQ.total <- na.omit(ASQ_df_final[,-c(4, 7:18,22)]) | |
| # # Fit the logistic regression model | |
| # model <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| # ChildAge_ASQ_months_allchildren_cent + | |
| # Child_Sex + | |
| # Depressive_Symptom_Severity, | |
| # data = new_ASQ.total, | |
| # family = binomial, | |
| # na.action = "na.exclude") | |
| # | |
| # ## Cook's distance | |
| # plot(model, which = 4, id.n = 5) | |
| # | |
| # #influential cases | |
| # # Extract model results | |
| # #new_ASQ <- ASQ_df_final[,-c(10:14)] #first exclude the other outcome variables | |
| # model.data <- augment(model, data = new_ASQ.total) %>% mutate(index = 1:n()) | |
| # model.data %>% top_n(5, .cooksd) | |
| # | |
| # #plot standardize residuls | |
| # ggplot(model.data, aes(index, .std.resid)) + | |
| # geom_point(aes(color = Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom), alpha = .5) + | |
| # theme_bw() | |
| # model.data %>% | |
| # filter(abs(.std.resid) > 3) #no influential cases | |
| # | |
| # #multicollinearity | |
| # car::vif(model) | |
| ## models | |
| M1_total_cat <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Depressive_Symptom_Severity, | |
| data = ASQ_df_final, | |
| family = "binomial") | |
| summary(M1_total_cat) | |
| res <- logit_apa(M1_total_cat,"Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| ASQ_total_results[6,8] <- res$estimate | |
| ASQ_total_results[6,9] <- res$SE | |
| ASQ_total_results[6,10] <- res$LL | |
| ASQ_total_results[6,11] <- res$UL | |
| ASQ_total_results[6,12] <- res$z | |
| ASQ_total_results[6,13] <- res$p | |
| #ASQ and prenatal Cesd*child sex | |
| M1_total_cat.INT1 <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Depressive_Symptom_Severity + | |
| Depressive_Symptom_Severity:Child_Sex, | |
| data = ASQ_df_final, | |
| family = "binomial") | |
| summary(M1_total_cat.INT1) | |
| logit.res_CESDxChS_cat <- logit_apa(M1_total_cat.INT1, "Child_Sexgirl:Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| #ASQ and prenatal Cesd*caseVScontrol | |
| M1_total_cat.INT2 <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| caseVScontrol + | |
| Depressive_Symptom_Severity + | |
| Depressive_Symptom_Severity:caseVScontrol, | |
| data = ASQ_df_final, | |
| family = "binomial") | |
| summary(M1_total_cat.INT2) | |
| logit.res_CESDxCaCo_cat <- logit_apa(M1_total_cat.INT2, "caseVScontrolcontrol:Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| #ASQ and prenatal Cesd + prenatal covariates | |
| M2_total_cat.cov <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Depressive_Symptom_Severity + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent, | |
| data = ASQ_df_final, | |
| family = "binomial") | |
| summary(M2_total_cat.cov) | |
| res <- logit_apa(M2_total_cat.cov,"Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| ASQ_total_results[7,8] <- res$estimate | |
| ASQ_total_results[7,9] <- res$SE | |
| ASQ_total_results[7,10] <- res$LL | |
| ASQ_total_results[7,11] <- res$UL | |
| ASQ_total_results[7,12] <- res$z | |
| ASQ_total_results[7,13] <- res$p | |
| #ASQ and prenatal Cesd + postnatal covariates | |
| M3_total_cat.cov <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Depressive_Symptom_Severity + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent + | |
| postpartum_Cesd_cent + | |
| postpartum_BAI_cent, | |
| data = ASQ_df_final, | |
| family = "binomial") | |
| summary(M3_total_cat.cov) | |
| res <- logit_apa(M3_total_cat.cov,"Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| ASQ_total_results[8,8] <- res$estimate | |
| ASQ_total_results[8,9] <- res$SE | |
| ASQ_total_results[8,10] <- res$LL | |
| ASQ_total_results[8,11] <- res$UL | |
| ASQ_total_results[8,12] <- res$z | |
| ASQ_total_results[8,13] <- res$p | |
| ## Additive CES-D effects #### | |
| ## models | |
| M1_total_cat_add <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD, | |
| data = ASQ_df_final, | |
| family = "binomial") | |
| summary(M1_total_cat_add ) | |
| ##document results | |
| logit_res.addD_pre <- logit_apa(M1_total_cat_add, "Pre.Post_clinDpre_only") | |
| logit_res.addD_post <- logit_apa(M1_total_cat_add, "Pre.Post_clinDpost_only") | |
| logit_res.addD_preANDpost <- logit_apa(M1_total_cat_add, "Pre.Post_clinDpre_post") | |
| M1_total_cat_add.vsPOST <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| relevel(Pre.Post_clinD, ref = "post_only"), | |
| data = ASQ_df_final, | |
| na.action = "na.exclude") | |
| #results relative to post-only | |
| logit_res.cat_addD_pre_post.VS.post_only <- logit_apa(M1_total_cat_add.vsPOST, "relevel(Pre.Post_clinD, ref = \"post_only\")pre_post") | |
| #ASQ and prenatal Cesd*child sex | |
| M1_total_cat.INT1_add1 <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD, | |
| data = ASQ_df_final, | |
| family = "binomial") | |
| summary(M1_total_cat.INT1_add1) | |
| M1_total_cat.INT1_add2 <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD + | |
| Pre.Post_clinD:Child_Sex, | |
| data = ASQ_df_final, | |
| family = "binomial") | |
| summary(M1_total_cat.INT1_add2) | |
| ##document results | |
| p <- pchisq(2 * (logLik(M1_total_cat.INT1_add1) - logLik(M1_total_cat.INT1_add2)), | |
| df = 2, | |
| lower.tail = FALSE) | |
| Logit_res.CESDxChS_add <- list(logLik(M1_total_cat.INT1_add1),logLik(M1_total_cat.INT1_add2), p) | |
| #ASQ and prenatal Cesd*caseVScontrol | |
| M1_total_cat.INT2_add1 <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| caseVScontrol + | |
| Pre.Post_clinD, | |
| data = ASQ_df_final, | |
| family = "binomial") | |
| summary(M1_total_cat.INT2_add1) | |
| M1_total_cat.INT2_add2 <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| caseVScontrol + | |
| Pre.Post_clinD + | |
| Pre.Post_clinD:caseVScontrol, | |
| data = ASQ_df_final, | |
| family = "binomial") | |
| summary(M1_total_cat.INT2_add2) | |
| ##document results | |
| p <- pchisq(2 * (logLik(M1_total_cat.INT2_add1) - logLik(M1_total_cat.INT2_add2)), | |
| df = 2, | |
| lower.tail = FALSE) | |
| Logit_res.CESDxCaCo_add <- list(logLik(M1_total_cat.INT2_add1),logLik(M1_total_cat.INT2_add2), p) | |
| #ASQ and prenatal Cesd + prenatal covariates | |
| M2_total_cat.cov_add <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent, | |
| data = ASQ_df_final, | |
| family = "binomial") | |
| summary(M2_total_cat.cov_add) | |
| #ASQ and prenatal Cesd + postnatal covariates | |
| M3_total_cat.cov_add <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent + | |
| postpartum_Cesd_cent + | |
| postpartum_BAI_cent, | |
| data = ASQ_df_final, | |
| family = "binomial") | |
| summary(M3_total_cat.cov_add) | |
| ############################################################################### | |
| # ## leave out this analysis??? additive effects of clinical depression #### | |
| # #test this only in participants assessed at least twice during pregnancy | |
| # # multiple_assessments_sub <- ASQ_df_final[ASQ_df_final$participantID %in% at_least_two_assessments & !is.na(ASQ_df_final$Cesd_qclosest_to_cortGW_pregMean_cent),] | |
| # # multiple_assessments_sub$additive_clinD <- factor(multiple_assessments_sub$additive_clinD, | |
| # # levels = c(0,1,2,3), | |
| # # labels = c("never","once","multiple", "multiple")) | |
| # # M.CESD_add <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| # # ChildAge_ASQ_months_allchildren_cent + | |
| # # Child_Sex + | |
| # # additive_clinD, | |
| # # tobit(Upper = 300), | |
| # # data = multiple_assessments_sub, | |
| # # na.action = "na.exclude") | |
| # # summary(M.CESD_add) | |
| # # | |
| # # ##document results | |
| # # tobit_res.addD_once <- tobit_apa(M.CESD_add, "additive_clinDonce") | |
| # # tobit_res.addD_multiple <- tobit_apa(M.CESD_add, "additive_clinDmultiple") | |
| # # | |
| # # ##test assumptions | |
| # # multiple_assessments_sub$yhat <- fitted(M.CESD_add)[,1] | |
| # # multiple_assessments_sub$rr <- resid(M.CESD_add, type = "response") | |
| # # multiple_assessments_sub$rp <- resid(M.CESD_add, type = "pearson")[,1] | |
| # # | |
| # # par(mfcol = c(2, 3)) | |
| # # | |
| # # with(multiple_assessments_sub, { | |
| # # plot(yhat, rr, main = "Fitted vs Residuals") | |
| # # qqnorm(rr) | |
| # # plot(yhat, rp, main = "Fitted vs Pearson Residuals") | |
| # # qqnorm(rp) | |
| # # plot(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange, | |
| # # rp, main = "Actual vs Pearson Residuals") | |
| # # plot(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange, | |
| # # yhat, main = "Actual vs Fitted") | |
| # # }) | |
| # # summary(M.CESD) | |
| # # | |
| # # #Sex-specific effects | |
| # # M.CESDINT_ChS_add <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| # # ChildAge_ASQ_months_allchildren_cent + | |
| # # Child_Sex + | |
| # # additive_clinD + | |
| # # additive_clinD:Child_Sex, | |
| # # tobit(Upper = 300), | |
| # # data = multiple_assessments_sub) | |
| # # summary(M.CESDINT_ChS_add) | |
| # # | |
| # # #caseVScontrol | |
| # # M.CESDINT_CaCo_add <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| # # ChildAge_ASQ_months_allchildren_cent + | |
| # # Child_Sex + | |
| # # caseVScontrol + | |
| # # additive_clinD + | |
| # # additive_clinD:caseVScontrol, | |
| # # tobit(Upper = 300), | |
| # # data = multiple_assessments_sub) | |
| # # summary(M.CESDINT_CaCo_add) | |
| # # | |
| # # ### adding prenatal covariates | |
| # # M.CESD.cov1_add <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| # # ChildAge_ASQ_months_allchildren_cent + | |
| # # Child_Sex + | |
| # # additive_clinD + | |
| # # #BAI_qclosest_to_cortGW_pregMean_cent + | |
| # # #PSQI_qclosest_to_cortGW_pregMean_cent + | |
| # # caseVScontrol + | |
| # # Maternal_Age_Years_cent + | |
| # # Maternal_Education + | |
| # # Parity + | |
| # # Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| # # Weight_Gain_cent + | |
| # # Maternal_Hypertensive_Disorders_anyVSnone + | |
| # # Maternal_Diabetes_Disorders_anyVSnone + | |
| # # Maternal_Smoking_During_Pregnancy + | |
| # # Gestational_Age_Weeks_cent + | |
| # # Child_Birth_Weight_cent, | |
| # # tobit(Upper = 300), | |
| # # data = multiple_assessments_sub) | |
| # # summary(M.CESD.cov1_add) | |
| # # | |
| # # ## adding the postnatal factors | |
| # # M.CESD.cov2_add <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| # # ChildAge_ASQ_months_allchildren_cent + | |
| # # Child_Sex + | |
| # # additive_clinD + | |
| # # #BAI_qclosest_to_cortGW_pregMean_cent + | |
| # # #PSQI_qclosest_to_cortGW_pregMean_cent + | |
| # # caseVScontrol + | |
| # # Maternal_Age_Years_cent + | |
| # # Maternal_Education + | |
| # # Parity + | |
| # # Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| # # Weight_Gain_cent + | |
| # # Maternal_Hypertensive_Disorders_anyVSnone + | |
| # # Maternal_Diabetes_Disorders_anyVSnone + | |
| # # Maternal_Smoking_During_Pregnancy + | |
| # # Gestational_Age_Weeks_cent + | |
| # # Child_Birth_Weight_cent + | |
| # # postpartum_Cesd_cent + | |
| # # postpartum_BAI_cent, | |
| # # tobit(Upper = 300), | |
| # # data = multiple_assessments_sub) | |
| # # summary(M.CESD.cov2_add) | |
| # | |
| # ## Additive CES-D effects #### | |
| # multiple_assessments_sub <- ASQ_df_final[ASQ_df_final$participantID %in% at_least_two_assessments & !is.na(ASQ_df_final$Cesd_qclosest_to_cortGW_pregMean_cent),] | |
| # multiple_assessments_sub$additive_clinD <- factor(multiple_assessments_sub$additive_clinD, | |
| # levels = c(0,1,2,3), | |
| # labels = c("never","once","multiple", "multiple")) | |
| # ## models | |
| # M1_total_cat_add <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| # ChildAge_ASQ_months_allchildren_cent + | |
| # Child_Sex + | |
| # additive_clinD, | |
| # data = multiple_assessments_sub, | |
| # #data = age_sub, | |
| # family = "binomial") | |
| # summary(M1_total_cat_add ) | |
| # | |
| # #ASQ and prenatal Cesd*child sex | |
| # M1_total_cat.INT1_add <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| # ChildAge_ASQ_months_allchildren_cent + | |
| # Child_Sex + | |
| # additive_clinD + | |
| # additive_clinD:Child_Sex, | |
| # data = multiple_assessments_sub, | |
| # family = "binomial") | |
| # summary(M1_total_cat.INT1) | |
| # | |
| # #ASQ and prenatal Cesd*caseVScontrol | |
| # M1_total_cat.INT2_add <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| # ChildAge_ASQ_months_allchildren_cent + | |
| # Child_Sex + | |
| # caseVScontrol + | |
| # additive_clinD + | |
| # additive_clinD:caseVScontrol, | |
| # data = multiple_assessments_sub, | |
| # family = "binomial") | |
| # summary(M1_total_cat.INT2_add ) | |
| # | |
| # #ASQ and prenatal Cesd + prenatal covariates | |
| # M2_total_cat.cov_add <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| # ChildAge_ASQ_months_allchildren_cent + | |
| # Child_Sex + | |
| # additive_clinD + | |
| # caseVScontrol + | |
| # Maternal_Age_Years_cent + | |
| # Maternal_Education + | |
| # Parity + | |
| # Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| # Weight_Gain_cent + | |
| # Maternal_Hypertensive_Disorders_anyVSnone + | |
| # Maternal_Diabetes_Disorders_anyVSnone + | |
| # Maternal_Smoking_During_Pregnancy + | |
| # Gestational_Age_Weeks_cent + | |
| # Child_Birth_Weight_cent, | |
| # data = multiple_assessments_sub, | |
| # family = "binomial") | |
| # summary(M2_total_cat.cov_add ) | |
| # | |
| # #ASQ and prenatal Cesd + postnatal covariates | |
| # M3_total_cat.cov_add <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| # ChildAge_ASQ_months_allchildren_cent + | |
| # Child_Sex + | |
| # additive_clinD + | |
| # caseVScontrol + | |
| # Maternal_Age_Years_cent + | |
| # Maternal_Education + | |
| # Parity + | |
| # Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| # Weight_Gain_cent + | |
| # Maternal_Hypertensive_Disorders_anyVSnone + | |
| # Maternal_Diabetes_Disorders_anyVSnone + | |
| # Maternal_Smoking_During_Pregnancy + | |
| # Gestational_Age_Weeks_cent + | |
| # Child_Birth_Weight_cent + | |
| # postpartum_Cesd_cent + | |
| # postpartum_BAI_cent, | |
| # data = multiple_assessments_sub, | |
| # family = "binomial") | |
| # summary(M3_total_cat.cov_add) | |
| ##########################3 Sensitivity analysis ######################### | |
| ## Set up results table #### | |
| Appen_ASQ_total_results <- setNames(data.frame(matrix(ncol = 13, nrow = 8)), | |
| c("Maternal depressive symptoms", | |
| "B", | |
| "SE", | |
| "LL", | |
| "UL", | |
| "z", | |
| "p", | |
| "OR", | |
| "SE", | |
| "LL", | |
| "UL", | |
| "z", | |
| "p")) | |
| Appen_ASQ_total_results[,1] <- c("Pregnancy mean CES-D score (n = ", | |
| "Model 1", | |
| "Model 2", | |
| "Model 3", | |
| "Pregnancy mean CES-D score >= 16", | |
| "Model 1", | |
| "Model 2", | |
| "Model 3") | |
| ################################################################################ | |
| #### Tobit on Pregnancy Averages #### | |
| ## CES-D #### | |
| M.CESD <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Cesd_qclosest_to_cortGW_pregMean_cent, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final_sub, | |
| na.action = "na.exclude") | |
| summary(M.CESD) | |
| res <- tobit_apa(M.CESD, "Cesd_qclosest_to_cortGW_pregMean_cent") | |
| Appen_ASQ_total_results[2,2] <- res$estimate | |
| Appen_ASQ_total_results[2,3] <- res$SE | |
| Appen_ASQ_total_results[2,4] <- res$LL | |
| Appen_ASQ_total_results[2,5] <- res$UL | |
| Appen_ASQ_total_results[2,6] <- res$z | |
| Appen_ASQ_total_results[2,7] <- res$p | |
| M.PSQI <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| PSQI_qclosest_to_cortGW_pregMean_cent, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final_sub, | |
| na.action = "na.exclude") | |
| summary(M.PSQI) | |
| #document results | |
| Appen_Tobit_res.PSQI <- tobit_apa(M.PSQI, "PSQI_qclosest_to_cortGW_pregMean_cent") | |
| M.BAI <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| BAI_qclosest_to_cortGW_pregMean_cent, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final_sub, | |
| na.action = "na.exclude") | |
| summary(M.BAI) | |
| #document results | |
| Appen_Tobit_res.BAI <- tobit_apa(M.BAI, "BAI_qclosest_to_cortGW_pregMean_cent") | |
| #Sex-specific effects | |
| M.CESDINT_ChS <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Cesd_qclosest_to_cortGW_pregMean_cent + | |
| Cesd_qclosest_to_cortGW_pregMean_cent:Child_Sex, | |
| tobit(Upper = 300), data = ASQ_df_final_sub) | |
| summary(M.CESDINT_ChS) | |
| #document results | |
| Appen_Tobit_res.CESDxChS <- tobit_apa(M.CESDINT_ChS, "Child_Sexgirl:Cesd_qclosest_to_cortGW_pregMean_cent") | |
| #caseVScontrol | |
| M.CESDINT_CaCo <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| caseVScontrol + | |
| Cesd_qclosest_to_cortGW_pregMean_cent + | |
| Cesd_qclosest_to_cortGW_pregMean_cent:caseVScontrol, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final_sub) | |
| summary(M.CESDINT_CaCo) | |
| #document results | |
| Appen_Tobit_res.CESDxCaCo <- tobit_apa(M.CESDINT_CaCo, "caseVScontrolcontrol:Cesd_qclosest_to_cortGW_pregMean_cent") | |
| ### adding prenatal covariates | |
| M.CESD.cov1 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Cesd_qclosest_to_cortGW_pregMean_cent + | |
| #BAI_qclosest_to_cortGW_pregMean_cent + | |
| #PSQI_qclosest_to_cortGW_pregMean_cent + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent, | |
| tobit(Upper = 300), | |
| na.action = "na.exclude", | |
| data = ASQ_df_final_sub) | |
| summary(M.CESD.cov1) | |
| #document results | |
| res <- tobit_apa(M.CESD.cov1, "Cesd_qclosest_to_cortGW_pregMean_cent") | |
| Appen_ASQ_total_results[3,2] <- res$estimate | |
| Appen_ASQ_total_results[3,3] <- res$SE | |
| Appen_ASQ_total_results[3,4] <- res$LL | |
| Appen_ASQ_total_results[3,5] <- res$UL | |
| Appen_ASQ_total_results[3,6] <- res$z | |
| Appen_ASQ_total_results[3,7] <- res$p | |
| #total variance explained by tobit model | |
| ASQ_df_final_sub$yhat <- fitted(M.CESD.cov1)[,1] | |
| r <- cor(ASQ_df_final_sub$yhat, | |
| ASQ_df_final_sub$Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange, | |
| use = "pairwise.complete.obs") | |
| Appen_Tobit_var.explained.prenat <- format(round(r^2,2)) | |
| ## adding the postnatal factors | |
| M.CESD.cov2 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Cesd_qclosest_to_cortGW_pregMean_cent + | |
| #BAI_qclosest_to_cortGW_pregMean_cent + | |
| #PSQI_qclosest_to_cortGW_pregMean_cent + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent + | |
| postpartum_Cesd_cent + | |
| postpartum_BAI_cent, | |
| tobit(Upper = 300), | |
| na.action = "na.exclude", | |
| data = ASQ_df_final_sub) | |
| summary(M.CESD.cov2) | |
| #document results | |
| res <- tobit_apa(M.CESD.cov2, "Cesd_qclosest_to_cortGW_pregMean_cent") | |
| Appen_ASQ_total_results[4,2] <- res$estimate | |
| Appen_ASQ_total_results[4,3] <- res$SE | |
| Appen_ASQ_total_results[4,4] <- res$LL | |
| Appen_ASQ_total_results[4,5] <- res$UL | |
| Appen_ASQ_total_results[4,6] <- res$z | |
| Appen_ASQ_total_results[4,7] <- res$p | |
| ## also report sig. covariates | |
| # COV.tobit_ChS <- tobit_apa(M.CESD.cov2, "Child_Sexgirl") | |
| # COV.tobit_Diab <- tobit_apa(M.CESD.cov2, "Maternal_Diabetes_Disorders_anyVSnoneyes") | |
| # COV.tobit_birthW <- tobit_apa(M.CESD.cov2, "Child_Birth_Weight_cent") | |
| # COV.tobit_postD <- tobit_apa(M.CESD.cov2, "postpartum_Cesd_cent") | |
| #total variance explained by tobit model | |
| ASQ_df_final_sub$yhat <- fitted(M.CESD.cov2)[,1] | |
| r <- cor(ASQ_df_final_sub$yhat, | |
| ASQ_df_final_sub$Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange, | |
| use = "pairwise.complete.obs") | |
| Appen_Tobit_var.explained_post <- r^2 | |
| ## categorical depression #### | |
| M.CESD_cat <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Depressive_Symptom_Severity, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final_sub) | |
| summary(M.CESD_cat) | |
| #document results | |
| res <- tobit_apa(M.CESD_cat, "Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| Appen_ASQ_total_results[6,2] <- res$estimate | |
| Appen_ASQ_total_results[6,3] <- res$SE | |
| Appen_ASQ_total_results[6,4] <- res$LL | |
| Appen_ASQ_total_results[6,5] <- res$UL | |
| Appen_ASQ_total_results[6,6] <- res$z | |
| Appen_ASQ_total_results[6,7] <- res$p | |
| ##INT childsex | |
| M.CESD_cat.INT1 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Depressive_Symptom_Severity + | |
| Depressive_Symptom_Severity:Child_Sex, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final_sub) | |
| summary(M.CESD_cat.INT1) | |
| #document results | |
| Appen_Tobit_res.CESDxChS_cat <- tobit_apa(M.CESD_cat.INT1, "Child_Sexgirl:Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| #INT caseVScontrol | |
| M.CESD_cat.INT2 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| caseVScontrol + | |
| Depressive_Symptom_Severity + | |
| Depressive_Symptom_Severity:caseVScontrol, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final_sub) | |
| summary(M.CESD_cat.INT2) | |
| #document | |
| Appen_Tobit_res.CESDxCaCo_cat <- tobit_apa(M.CESD_cat.INT2, "caseVScontrolcontrol:Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| ## incl. prenatal covariates | |
| M.CESD_cat.cov1 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Depressive_Symptom_Severity + | |
| #BAI_qclosest_to_cortGW_pregMean_cent + | |
| #PSQI_qclosest_to_cortGW_pregMean_cent + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final_sub) | |
| summary(M.CESD_cat.cov1) | |
| #document results | |
| res <- tobit_apa(M.CESD_cat.cov1, "Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| Appen_ASQ_total_results[7,2] <- res$estimate | |
| Appen_ASQ_total_results[7,3] <- res$SE | |
| Appen_ASQ_total_results[7,4] <- res$LL | |
| Appen_ASQ_total_results[7,5] <- res$UL | |
| Appen_ASQ_total_results[7,6] <- res$z | |
| Appen_ASQ_total_results[7,7] <- res$p | |
| ## incl. postpartum covariates | |
| M.CESD_cat.cov2 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Depressive_Symptom_Severity + | |
| #BAI_qclosest_to_cortGW_pregMean_cent + | |
| #PSQI_qclosest_to_cortGW_pregMean_cent + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent + | |
| postpartum_BAI_cent + | |
| postpartum_Cesd_cent, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final_sub) | |
| summary(M.CESD_cat.cov2) | |
| ##document results | |
| res <- tobit_apa(M.CESD_cat.cov2, "Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| Appen_ASQ_total_results[8,2] <- res$estimate | |
| Appen_ASQ_total_results[8,3] <- res$SE | |
| Appen_ASQ_total_results[8,4] <- res$LL | |
| Appen_ASQ_total_results[8,5] <- res$UL | |
| Appen_ASQ_total_results[8,6] <- res$z | |
| Appen_ASQ_total_results[8,7] <- res$p | |
| ## additive pre-post clinical depression #### | |
| M.CESD_add <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final_sub, | |
| na.action = "na.exclude") | |
| summary(M.CESD_add) | |
| ##document results | |
| Appen_tobit_res.addD_pre <- tobit_apa(M.CESD_add, "Pre.Post_clinDpre_only") | |
| Appen_tobit_res.addD_post <- tobit_apa(M.CESD_add, "Pre.Post_clinDpost_only") | |
| Appen_tobit_res.addD_preANDpost <- tobit_apa(M.CESD_add, "Pre.Post_clinDpre_post") | |
| M.CESD_add.vsPOST <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| relevel(Pre.Post_clinD, ref = "post_only"), | |
| tobit(Upper = 300), | |
| data = ASQ_df_final_sub, | |
| na.action = "na.exclude") | |
| rownames(coef(summary(M.CESD_add.vsPOST))) | |
| #results relative to post-only | |
| Appen_tobit_res.addD_pre_post.VS.post_only <- tobit_apa(M.CESD_add.vsPOST, "relevel(Pre.Post_clinD, ref = \"post_only\")pre_post") | |
| summary(M.CESD) | |
| #Sex-specific effects | |
| M.CESDINT_ChS_add1 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final_sub) | |
| M.CESDINT_ChS_add2 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD + | |
| Pre.Post_clinD:Child_Sex, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final_sub) | |
| p <- pchisq(2 * (logLik(M.CESDINT_ChS_add1) - logLik(M.CESDINT_ChS_add2)), | |
| df = 2, | |
| lower.tail = FALSE) | |
| Appen_Tobit_res.CESDxChS_add <- list(logLik(M.CESDINT_ChS_add1),logLik(M.CESDINT_ChS_add2), p) | |
| #caseVScontrol | |
| #without interaction | |
| M.CESDINT_CaCo_add1 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| caseVScontrol + | |
| Pre.Post_clinD, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final_sub) | |
| summary(M.CESDINT_CaCo_add1) | |
| #with interaction | |
| M.CESDINT_CaCo_add2 <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| caseVScontrol + | |
| Pre.Post_clinD + | |
| Pre.Post_clinD:caseVScontrol, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final_sub) | |
| summary(M.CESDINT_CaCo_add2) | |
| ##document results | |
| p <- pchisq(2 * (logLik(M.CESDINT_CaCo_add1) - logLik(M.CESDINT_CaCo_add2)), | |
| df = 2, | |
| lower.tail = FALSE) | |
| Appen_Tobit_res.CESDxCaCo_add <- list(logLik(M.CESDINT_CaCo_add1),logLik(M.CESDINT_CaCo_add2), p) | |
| ### adding prenatal covariates | |
| M.CESD.cov1_add <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD + | |
| #BAI_qclosest_to_cortGW_pregMean_cent + | |
| #PSQI_qclosest_to_cortGW_pregMean_cent + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final_sub) | |
| summary(M.CESD.cov1_add) | |
| ## adding the postnatal factors | |
| M.CESD.cov2_add <- vglm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD + | |
| #BAI_qclosest_to_cortGW_pregMean_cent + | |
| #PSQI_qclosest_to_cortGW_pregMean_cent + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent + | |
| postpartum_BAI_cent, | |
| tobit(Upper = 300), | |
| data = ASQ_df_final_sub) | |
| summary(M.CESD.cov2_add) | |
| ################################################################################ | |
| ###### Logistic Regression on Neurodevelopmental delay #### | |
| ## Continuous CES-D Models #### | |
| #ASQ and prenatal Cesd | |
| M1_total <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Cesd_qclosest_to_cortGW_pregMean_cent, | |
| data = ASQ_df_final_sub, | |
| family = "binomial") | |
| summary(M1_total) | |
| res <- logit_apa(M1_total,"Cesd_qclosest_to_cortGW_pregMean_cent") | |
| Appen_ASQ_total_results[2,8] <- res$estimate | |
| Appen_ASQ_total_results[2,9] <- res$SE | |
| Appen_ASQ_total_results[2,10] <- res$LL | |
| Appen_ASQ_total_results[2,11] <- res$UL | |
| Appen_ASQ_total_results[2,12] <- res$z | |
| Appen_ASQ_total_results[2,13] <- res$p | |
| #ASQ and BAI | |
| M1_total_BAI <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| BAI_qclosest_to_cortGW_pregMean_cent, | |
| data = ASQ_df_final_sub, | |
| family = "binomial") | |
| summary(M1_total_BAI) | |
| Appen_logit.res_BAI <- logit_apa(M1_total_BAI,"BAI_qclosest_to_cortGW_pregMean_cent") | |
| #ASQ and PSQI | |
| M1_total_PSQI <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| PSQI_qclosest_to_cortGW_pregMean_cent, | |
| data = ASQ_df_final_sub, | |
| family = "binomial") | |
| summary(M1_total_PSQI) | |
| Appen_logit.res_PSQI <- logit_apa(M1_total_PSQI,"PSQI_qclosest_to_cortGW_pregMean_cent") | |
| #ASQ and prenatal Cesd*child sex | |
| M1_total.INT1 <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Cesd_qclosest_to_cortGW_pregMean_cent + | |
| Cesd_qclosest_to_cortGW_pregMean_cent:Child_Sex, | |
| data = ASQ_df_final_sub, | |
| family = "binomial") | |
| summary(M1_total.INT1) | |
| Appen_logit.res_CESDxChS <- logit_apa(M1_total.INT1,"Child_Sexgirl:Cesd_qclosest_to_cortGW_pregMean_cent") | |
| #ASQ and prenatal Cesd*caseVScontrol | |
| M1_total.INT2 <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| caseVScontrol + | |
| Cesd_qclosest_to_cortGW_pregMean_cent + | |
| Cesd_qclosest_to_cortGW_pregMean_cent:caseVScontrol, | |
| data = ASQ_df_final_sub, | |
| family = "binomial") | |
| summary(M1_total.INT2) | |
| Appen_logit.res_CESDxCaCo <- logit_apa(M1_total.INT2,"caseVScontrolcontrol:Cesd_qclosest_to_cortGW_pregMean_cent") | |
| #ASQ and prenatal Cesd + prenatal covariates | |
| M2_total.cov <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Cesd_qclosest_to_cortGW_pregMean_cent + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent, | |
| data = ASQ_df_final_sub, | |
| family = "binomial") | |
| summary(M2_total.cov) | |
| #document | |
| res <- logit_apa(M2_total.cov,"Cesd_qclosest_to_cortGW_pregMean_cent") | |
| Appen_ASQ_total_results[3,8] <- res$estimate | |
| Appen_ASQ_total_results[3,9] <- res$SE | |
| Appen_ASQ_total_results[3,10] <- res$LL | |
| Appen_ASQ_total_results[3,11] <- res$UL | |
| Appen_ASQ_total_results[3,12] <- res$z | |
| Appen_ASQ_total_results[3,13] <- res$p | |
| #ASQ and prenatal Cesd + postnatal covariates | |
| M3_total.cov <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Cesd_qclosest_to_cortGW_pregMean_cent + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent + | |
| postpartum_BAI_cent + | |
| postpartum_Cesd_cent, | |
| data = ASQ_df_final_sub, | |
| family = "binomial") | |
| summary(M3_total.cov) | |
| # COV.logit_maternal_edu <- logit_apa(M3_total.cov, "Maternal_Educationuniversity") | |
| # COV.logit_birthW <- logit_apa(M3_total.cov, "Child_Birth_Weight_cent") | |
| #document | |
| res <- logit_apa(M3_total.cov,"Cesd_qclosest_to_cortGW_pregMean_cent") | |
| Appen_ASQ_total_results[4,8] <- res$estimate | |
| Appen_ASQ_total_results[4,9] <- res$SE | |
| Appen_ASQ_total_results[4,10] <- res$LL | |
| Appen_ASQ_total_results[4,11] <- res$UL | |
| Appen_ASQ_total_results[4,12] <- res$z | |
| Appen_ASQ_total_results[4,13] <- res$p | |
| ## Categorical CES-D Models ##### | |
| # assumptions | |
| # new_ASQ.total <- na.omit(ASQ_df_final[,-c(4, 7:18,22)]) | |
| # # Fit the logistic regression model | |
| # model <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| # ChildAge_ASQ_months_allchildren_cent + | |
| # Child_Sex + | |
| # Depressive_Symptom_Severity, | |
| # data = new_ASQ.total, | |
| # family = binomial, | |
| # na.action = "na.exclude") | |
| # | |
| # ## Cook's distance | |
| # plot(model, which = 4, id.n = 5) | |
| # | |
| # #influential cases | |
| # # Extract model results | |
| # #new_ASQ <- ASQ_df_final[,-c(10:14)] #first exclude the other outcome variables | |
| # model.data <- augment(model, data = new_ASQ.total) %>% mutate(index = 1:n()) | |
| # model.data %>% top_n(5, .cooksd) | |
| # | |
| # #plot standardize residuls | |
| # ggplot(model.data, aes(index, .std.resid)) + | |
| # geom_point(aes(color = Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom), alpha = .5) + | |
| # theme_bw() | |
| # model.data %>% | |
| # filter(abs(.std.resid) > 3) #no influential cases | |
| # | |
| # #multicollinearity | |
| # car::vif(model) | |
| ## models | |
| M1_total_cat <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Depressive_Symptom_Severity, | |
| data = ASQ_df_final_sub, | |
| family = "binomial") | |
| summary(M1_total_cat) | |
| res <- logit_apa(M1_total_cat,"Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| Appen_ASQ_total_results[6,8] <- res$estimate | |
| Appen_ASQ_total_results[6,9] <- res$SE | |
| Appen_ASQ_total_results[6,10] <- res$LL | |
| Appen_ASQ_total_results[6,11] <- res$UL | |
| Appen_ASQ_total_results[6,12] <- res$z | |
| Appen_ASQ_total_results[6,13] <- res$p | |
| #ASQ and prenatal Cesd*child sex | |
| M1_total_cat.INT1 <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Depressive_Symptom_Severity + | |
| Depressive_Symptom_Severity:Child_Sex, | |
| data = ASQ_df_final_sub, | |
| family = "binomial") | |
| summary(M1_total_cat.INT1) | |
| Appen_logit.res_CESDxChS_cat <- logit_apa(M1_total_cat.INT1, "Child_Sexgirl:Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| #ASQ and prenatal Cesd*caseVScontrol | |
| M1_total_cat.INT2 <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| caseVScontrol + | |
| Depressive_Symptom_Severity + | |
| Depressive_Symptom_Severity:caseVScontrol, | |
| data = ASQ_df_final_sub, | |
| family = "binomial") | |
| summary(M1_total_cat.INT2) | |
| Appen_logit.res_CESDxCaCo_cat <- logit_apa(M1_total_cat.INT2, "caseVScontrolcontrol:Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| #ASQ and prenatal Cesd + prenatal covariates | |
| M2_total_cat.cov <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Depressive_Symptom_Severity + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent, | |
| data = ASQ_df_final_sub, | |
| family = "binomial") | |
| summary(M2_total_cat.cov) | |
| res <- logit_apa(M2_total_cat.cov,"Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| Appen_ASQ_total_results[7,8] <- res$estimate | |
| Appen_ASQ_total_results[7,9] <- res$SE | |
| Appen_ASQ_total_results[7,10] <- res$LL | |
| Appen_ASQ_total_results[7,11] <- res$UL | |
| Appen_ASQ_total_results[7,12] <- res$z | |
| Appen_ASQ_total_results[7,13] <- res$p | |
| #ASQ and prenatal Cesd + postnatal covariates | |
| M3_total_cat.cov <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Depressive_Symptom_Severity + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent + | |
| postpartum_Cesd_cent + | |
| postpartum_BAI_cent, | |
| data = ASQ_df_final_sub, | |
| family = "binomial") | |
| summary(M3_total_cat.cov) | |
| res <- logit_apa(M3_total_cat.cov,"Depressive_Symptom_SeverityClinical (CES-D >= 16)") | |
| Appen_ASQ_total_results[8,8] <- res$estimate | |
| Appen_ASQ_total_results[8,9] <- res$SE | |
| Appen_ASQ_total_results[8,10] <- res$LL | |
| Appen_ASQ_total_results[8,11] <- res$UL | |
| Appen_ASQ_total_results[8,12] <- res$z | |
| Appen_ASQ_total_results[8,13] <- res$p | |
| ## Additive CES-D effects #### | |
| ## models | |
| M1_total_cat_add <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD, | |
| data = ASQ_df_final_sub, | |
| family = "binomial") | |
| summary(M1_total_cat_add ) | |
| ##document results | |
| Appen_logit_res.addD_pre <- logit_apa(M1_total_cat_add, "Pre.Post_clinDpre_only") | |
| Appen_logit_res.addD_post <- logit_apa(M1_total_cat_add, "Pre.Post_clinDpost_only") | |
| Appen_logit_res.addD_preANDpost <- logit_apa(M1_total_cat_add, "Pre.Post_clinDpre_post") | |
| M1_total_cat_add.vsPOST <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| relevel(Pre.Post_clinD, ref = "post_only"), | |
| data = ASQ_df_final_sub, | |
| na.action = "na.exclude") | |
| #results relative to post-only | |
| Appen_logit_res.cat_addD_pre_post.VS.post_only <- logit_apa(M1_total_cat_add.vsPOST, "relevel(Pre.Post_clinD, ref = \"post_only\")pre_post") | |
| #ASQ and prenatal Cesd*child sex | |
| M1_total_cat.INT1_add1 <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD, | |
| data = ASQ_df_final_sub, | |
| family = "binomial") | |
| summary(M1_total_cat.INT1_add1) | |
| M1_total_cat.INT1_add2 <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD + | |
| Pre.Post_clinD:Child_Sex, | |
| data = ASQ_df_final_sub, | |
| family = "binomial") | |
| summary(M1_total_cat.INT1_add2) | |
| ##document results | |
| p <- pchisq(2 * (logLik(M1_total_cat.INT1_add1) - logLik(M1_total_cat.INT1_add2)), | |
| df = 2, | |
| lower.tail = FALSE) | |
| Appen_Logit_res.CESDxChS_add <- list(logLik(M1_total_cat.INT1_add1),logLik(M1_total_cat.INT1_add2), p) | |
| #ASQ and prenatal Cesd*caseVScontrol | |
| M1_total_cat.INT2_add1 <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| caseVScontrol + | |
| Pre.Post_clinD, | |
| data = ASQ_df_final_sub, | |
| family = "binomial") | |
| summary(M1_total_cat.INT2_add1) | |
| M1_total_cat.INT2_add2 <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| caseVScontrol + | |
| Pre.Post_clinD + | |
| Pre.Post_clinD:caseVScontrol, | |
| data = ASQ_df_final_sub, | |
| family = "binomial") | |
| summary(M1_total_cat.INT2_add2) | |
| ##document results | |
| p <- pchisq(2 * (logLik(M1_total_cat.INT2_add1) - logLik(M1_total_cat.INT2_add2)), | |
| df = 2, | |
| lower.tail = FALSE) | |
| Appen_Logit_res.CESDxCaCo_add <- list(logLik(M1_total_cat.INT2_add1),logLik(M1_total_cat.INT2_add2), p) | |
| #ASQ and prenatal Cesd + prenatal covariates | |
| M2_total_cat.cov_add <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent, | |
| data = ASQ_df_final_sub, | |
| family = "binomial") | |
| summary(M2_total_cat.cov_add) | |
| #ASQ and prenatal Cesd + postnatal covariates | |
| M3_total_cat.cov_add <- glm(Child_ASQ_totaldevelopmentalmilestones_Infancy_Sum_finalagerange_norm_dichom ~ | |
| ChildAge_ASQ_months_allchildren_cent + | |
| Child_Sex + | |
| Pre.Post_clinD + | |
| caseVScontrol + | |
| Maternal_Age_Years_cent + | |
| Maternal_Education + | |
| Parity + | |
| Maternal_Body_Mass_Index_in_Early_Pregnancy_cent + | |
| Weight_Gain_cent + | |
| Maternal_Hypertensive_Disorders_anyVSnone + | |
| Maternal_Diabetes_Disorders_anyVSnone + | |
| Maternal_Smoking_During_Pregnancy + | |
| Gestational_Age_Weeks_cent + | |
| Child_Birth_Weight_cent + | |
| postpartum_Cesd_cent + | |
| postpartum_BAI_cent, | |
| data = ASQ_df_final_sub, | |
| family = "binomial") | |
| summary(M3_total_cat.cov_add) |