NTD Study

Read in data

df_ntd_cases <- 
    readxl::read_xlsx("ntd-data.xlsx", sheet = 1) %>% 
    janitor::clean_names() %>% 
    mutate(
        case_control = rep("Case", by = nrow(.)),
        id = row_number(),
        presc = as.character(presc))

df_ntd_control <- 
    readxl::read_xlsx("ntd-data.xlsx", sheet = 2) %>% 
    janitor::clean_names() %>% 
    drop_na(bw) %>% 
    mutate(
        case_control = rep("Control", by = nrow(.)),
        id = row_number()+100,
        fa_start = as.character(fa_start),
        prev_ano = as.character(prev_ano),
        booking = as.character(booking))

df_ntd <-
    df_ntd_cases %>% 
    full_join(df_ntd_control)

df_ntd <- 
    df_ntd %>% 
    rename(
        bwt = bw,
        gestage = ga) %>% 
    mutate(
        sex = toupper(sex) %>% 
            factor(
                levels = c("F", "M"), 
                labels = c("Female", "Male")),
        type_gest = factor(
            type_gest, 
            levels = c("M","S"), c("Multiple", "Singleton")),
        mode_del = factor(mode_del),
        prev_ano = parse_integer(prev_ano),
        across(
            c(consanguinity, feb_ill, rash, herbal, alcohol, illicit_drugs, 
              anc_dx, smokin, prev_ano), 
            ~factor(.x, levels = c(0,1), labels = c("No", "Yes"))),
        case_control = factor(case_control),
        edu_level = stringr::str_to_sentence(edu_level) %>% 
            factor(),
        booking = parse_integer(booking),
        fa_start = parse_integer(fa_start),
        anc_attendance = case_when(
            anc_att == "non-att" ~ "Non-Attendant",
            anc_att == "non Attend" ~ "Non-Attendant",
            anc_att == "reg" ~ "Regular Attendant") %>% 
            factor(),
        anencephaly = case_when(
            anoma_ac1_ac34ly == "anencephaly" ~ "Yes", 
            case_control == "Case" ~ "No"),
        myelomen = case_when(
            str_detect(anoma_ac1_ac34ly, "MM") ~ "Yes", 
            case_control == "Case" ~ "No"),
        encephalocele = case_when(
            str_detect(anoma_ac1_ac34ly, "locele") ~ "Yes", 
            case_control == "Case" ~ "No"),
        hydroceph = case_when(
            str_detect(anoma_ac1_ac34ly, "HC") ~ "Yes", 
            case_control == "Case" ~ "No"),
        presc = case_when(presc == "0" ~ "No", TRUE ~ "Yes"),
        resid_mining = case_when(residence == "M" ~ "Yes", TRUE ~ "No"),
        resid_farming = case_when(residence == "F" ~ "Yes", TRUE ~ "No"),
        mat_dx = case_when(mat_dx == "0" ~ "No", TRUE ~ "Yes"),
        residence = factor(
            residence, 
            levels = c("C", "F", "M"),
            labels = c("None", "Farming","Mining")),
        case_control = factor(case_control, levels = c("Control", "Case"))
        ) %>% 
    select(-c(anc_att))

labelled::var_label(df_ntd) <-
    list(
        bwt = "Birth weight",
        gestage = "Gestational age",
        sex = "Sex",
        type_gest = "Gestation type",
        mode_del = "Delivery mode",
        consanguinity = "Consanguinity",
        mat_age = "Mother's age",
        edu_level = "Educational Level",
        booking = "Age at Booking ",
        anc_attendance = "ANC attendance",
        id = "Generate study ID",
        case_control = "Case or Control",
        parity = "Parity",
        herbal = "Herbal drug use",
        rash = "Rash",
        fa_start = "Folic Acid start",
        feb_ill = "Febirle illness",
        prev_ano = "Previous Anomalies",
        smokin = "Smoking",
        alcohol = "Alcohol use",
        illicit_drugs = "Illicit drug use",
        encephalocele = "Encephalocele",
        hydroceph = "Hydrocephalus",
        myelomen = "Myelomeningocele",
        anencephaly = "Anencephaly",
        unpres = "Unprescribed meds used",
        presc = "Prescribed meds use",
        resid_mining = "Mining area residence",
        resid_farming = "Farming area residence",
        mat_dx = "Maternal disease",
        residence = "Mining or Farming")

Table

gtsummary::theme_gtsummary_compact()

Setting theme "Compact"

table_1 <- 
    df_ntd %>% 
    select(
        bwt, gestage, sex, type_gest, mode_del, consanguinity, booking,
        mat_age, edu_level, feb_ill, rash, fa_start, herbal, mat_dx, 
        unpres, presc, smokin, alcohol,illicit_drugs, prev_ano, residence,
        resid_mining, resid_farming, anc_attendance, parity, 
        anc_dx, case_control) %>% 
    gtsummary::tbl_summary(
        by = case_control,
        type = list(parity ~ "continuous"),
        digits = list(
            gtsummary::all_continuous() ~ 1,
            gtsummary::all_categorical() ~ c(0,1),
            c(mat_age, gestage) ~ 0),
        statistic = list(gtsummary::all_categorical()~ "{n} ({p})")
        ) %>%
    gtsummary::add_overall(last =TRUE) %>% 
    gtsummary::add_p(
        pvalue_fun = function(x) gtsummary::style_pvalue(x, digits = 3)) %>% 
    gtsummary::bold_p() %>% 
    gtsummary::bold_labels()

table_1

Characteristic	Control N = 196 1	Case N = 97 1	Overall N = 293 1	p-value 2
Birth weight	2.9 (2.4, 3.4)	2.8 (2.5, 3.2)	2.9 (2.4, 3.4)	0.544
Gestational age	38 (36, 39)	38 (37, 39)	38 (36, 39)	0.955
Sex				0.248
Female	85 (43.4)	49 (50.5)	134 (45.7)
Male	111 (56.6)	48 (49.5)	159 (54.3)
Gestation type				0.091
Multiple	22 (11.2)	5 (5.2)	27 (9.2)
Singleton	174 (88.8)	92 (94.8)	266 (90.8)
Delivery mode				0.619
C/S	97 (49.5)	51 (52.6)	148 (50.5)
SVD	99 (50.5)	46 (47.4)	145 (49.5)
Consanguinity	0 (0.0)	1 (1.0)	1 (0.3)	0.331
Age at Booking	8.0 (6.5, 11.0)	12.0 (8.0, 16.0)	10.0 (7.0, 12.0)	<0.001
Unknown	0	4	4
Mother's age	30 (26, 33)	29 (26, 33)	30 (26, 33)	0.798
Educational Level				<0.001
None	9 (4.6)	19 (19.6)	28 (9.6)
Primary	89 (45.4)	35 (36.1)	124 (42.3)
Secondary	44 (22.4)	21 (21.6)	65 (22.2)
Tertiary	54 (27.6)	22 (22.7)	76 (25.9)
Febirle illness	2 (1.0)	5 (5.2)	7 (2.4)	0.042
Rash	0 (0.0)	0 (0.0)	0 (0.0)	>0.999
Folic Acid start	8.0 (6.5, 11.5)	12.0 (8.0, 17.0)	9.0 (7.0, 12.0)	<0.001
Unknown	0	4	4
Herbal drug use	13 (6.6)	33 (34.0)	46 (15.7)	<0.001
Maternal disease	20 (10.2)	9 (9.3)	29 (9.9)	0.803
Unprescribed meds used	4 (2.0)	8 (8.2)	12 (4.1)	0.023
Prescribed meds use	6 (3.1)	7 (7.2)	13 (4.4)	0.132
Smoking	0 (0.0)	0 (0.0)	0 (0.0)	>0.999
Alcohol use	0 (0.0)	1 (1.0)	1 (0.3)	0.331
Illicit drug use	0 (0.0)	0 (0.0)	0 (0.0)	>0.999
Previous Anomalies	0 (0.0)	1 (1.2)	1 (0.4)	0.302
Unknown	0	12	12
Mining or Farming				<0.001
None	172 (87.8)	62 (63.9)	234 (79.9)
Farming	9 (4.6)	22 (22.7)	31 (10.6)
Mining	15 (7.7)	13 (13.4)	28 (9.6)
Mining area residence	15 (7.7)	13 (13.4)	28 (9.6)	0.115
Farming area residence	9 (4.6)	22 (22.7)	31 (10.6)	<0.001
ANC attendance				0.017
Non-Attendant	3 (1.5)	7 (7.2)	10 (3.4)
Regular Attendant	193 (98.5)	90 (92.8)	283 (96.6)
Parity	3.0 (2.0, 3.0)	2.0 (1.0, 4.0)	2.0 (2.0, 3.0)	0.576
anc_dx	0 (0.0)	30 (30.9)	30 (10.2)	<0.001
1 Median (Q1, Q3); n (%)
2 Wilcoxon rank sum test; Pearson’s Chi-squared test; Fisher’s exact test

file.remove("ntd_table_1.docx")

[1] TRUE

table_1 %>% 
    gtsummary::as_gt() %>% 
    gt::gtsave(filename = "ntd_table_1.docx") 

Multivariate table

df_ntd %>% 
    glm(
        case_control ~ edu_level + feb_ill + fa_start + herbal + 
            unpres + resid_farming + anc_attendance, 
        family="binomial",
        data = .) %>% 
    gtsummary::tbl_regression(
        exponentiate = T,
        pvalue_fun = function(x) gtsummary::style_pvalue(x, digits = 3)
    ) %>%
    gtsummary::bold_labels() %>% 
    gtsummary::bold_p()

Characteristic	OR 1	95% CI 1	p-value
Educational Level
None	—	—
Primary	0.46	0.14, 1.48	0.189
Secondary	0.64	0.18, 2.28	0.483
Tertiary	0.89	0.26, 3.15	0.850
Febirle illness
No	—	—
Yes	3.46	0.50, 29.6	0.211
Folic Acid start	1.16	1.09, 1.23	<0.001
Herbal drug use
No	—	—
Yes	4.25	1.89, 9.84	<0.001
Unprescribed meds used	1.70	0.33, 8.52	0.509
Farming area residence
No	—	—
Yes	3.66	1.30, 10.6	0.015
ANC attendance
Non-Attendant	—	—
Regular Attendant	1.86	0.22, 18.0	0.577
1 OR = Odds Ratio, CI = Confidence Interval