Depressive Disorder Project
Thien
2023-11-24
1. Import library
suppressMessages(library(haven)) # for reading SAS file
suppressMessages(library(tidyverse))
suppressMessages(library(table1))
suppressMessages(library(compareGroups))
suppressMessages(library(epiDisplay)) # Summary the results of logistic regression
suppressMessages(library(GGally)) # Plot forest results
suppressMessages(library(ggplot2))
suppressMessages(library(ggthemes))2. Import data
df <- read_sas("C:/Users/thien/OneDrive/Desktop/SAS learning/nhanes.sas7bdat")3. Dependent variables (Depressive disorder)
table(df$DPQ010)##
## 0 1 2 3 7 9
## 3976 881 293 242 1 5table(df$DPQ020)##
## 0 1 2 3 7 9
## 4071 903 214 202 3 3table(df$DPQ030)##
## 0 1 2 3 7
## 3416 1127 350 501 1table(df$DPQ040)##
## 0 1 2 3 7
## 2650 1829 423 492 1table(df$DPQ050)##
## 0 1 2 3 9
## 4036 832 278 247 2table(df$DPQ060)##
## 0 1 2 3 9
## 4449 620 154 167 4table(df$DPQ070)##
## 0 1 2 3 7 9
## 4423 586 193 189 1 2table(df$DPQ080)##
## 0 1 2 3 7 9
## 4786 367 134 104 1 2table(df$DPQ090)##
## 0 1 2 3 7 9
## 5205 122 31 32 1 2# Filter the dataframe
df1 <- df %>%
filter(DPQ010 != 7 & DPQ010 != 9 & !is.na(DPQ010))
df2 <- df1 %>%
filter(DPQ020 != 7 & DPQ020 != 9 & !is.na(DPQ020))
df3 <- df2 %>%
filter(DPQ030 != 7 & DPQ030 != 9 & !is.na(DPQ030))
df4 <- df3 %>%
filter(DPQ040 != 7 & DPQ040 != 9 & !is.na(DPQ040))
df5 <- df4 %>%
filter(DPQ050 != 7 & DPQ050 != 9 & !is.na(DPQ050))
df6 <- df5 %>%
filter(DPQ060 != 7 & DPQ060 != 9 & !is.na(DPQ060))
df7 <- df6 %>%
filter(DPQ070 != 7 & DPQ070 != 9 & !is.na(DPQ070))
df8 <- df7 %>%
filter(DPQ080 != 7 & DPQ080 != 9 & !is.na(DPQ080))
df9 <- df8 %>%
filter(DPQ090 != 7 & DPQ090 != 9 & !is.na(DPQ090))
# Create depression variable:
depress <- df9
depress$dp = depress$DPQ010 + depress$DPQ020 + depress$DPQ030 + depress$DPQ040 + depress$DPQ050 +
depress$DPQ060 + depress$DPQ070 + depress$DPQ080 + depress$DPQ090
# Create the new variable 'dp10'
depress <- depress %>%
mutate(dp10 = ifelse(dp <= 9, 0, 1))
# Display frequency table for dp10
table(depress$dp10)##
## 0 1
## 4861 5114. Potential covariates
4.1 Create SBP and DBP
depress$SBP = (depress$BPXSY1 + depress$BPXSY2) / 2
depress$DBP =(depress$BPXDI1 + depress$BPXDI2) / 24.2 Create the variable ‘hyper’
hypertension <- depress %>%
mutate(hyper = case_when(
SBP >= 140 | DBP >= 90 | BPQ040A == 1 | BPQ050A == 1 ~ 1,
TRUE ~ 0))
table(hypertension$hyper)##
## 0 1
## 3458 19144.3 Create the variable ‘dyslip’ using case_when
lipid <- hypertension %>%
mutate(dyslip = case_when(
LBXTC >= 200 | LBXTR >= 150 | LBDLDL >= 100 | LBDHDD <= 40 | BPQ090D == 1 ~ 1,
TRUE ~ 0
))
# Display frequency table for 'dyslip'
table(lipid$dyslip)##
## 0 1
## 1681 36914.4 Create the ‘db’ variable using case_when
diab <- lipid %>%
mutate(db = case_when(
LBXGLU >= 126 | LBXGLT >= 200 | LBXGH >= 6.5 | DIQ070 == 1 ~ 1,
TRUE ~ 0
))
# Display frequency table for 'db'
table(diab$db)##
## 0 1
## 4542 830Select specific variables
# Selecting specific variables and creating a subset
subset_df <- subset(diab, select = c(RIDAGEYR, RIAGENDR, RIDRETH3, DMDEDUC2, DMDMARTL, SBP, DBP, BMXBMI, INDFMIN2, INDFMPIR, LBXSBU, LBXSCR, LBXSASSI, LBXSATSI, LBXAPB, LBDHDD, LBDLDL, LBXTR, LBXTC, LBXGLU, LBXGLT, ALQ101, ALQ110, SMQ020, PAQ665, PAQ650, LBXCOT, LBXHCT, URXCOTT, URXHCTT, db, dyslip, hyper, dp10))
selected_cols <- c("RIDAGEYR", "RIAGENDR", "RIDRETH3", "DMDEDUC2", "DMDMARTL", "SBP", "DBP",
"BMXBMI", "INDFMIN2", "INDFMPIR", "LBXSBU", "LBXSCR", "LBXSASSI", "LBXSATSI",
"LBXAPB", "LBDHDD", "LBDLDL", "LBXTR", "LBXTC", "LBXGLU", "LBXGLT",
"ALQ101", "ALQ110", "SMQ020", "PAQ665", "PAQ650", "LBXCOT", "LBXHCT",
"URXCOTT", "URXHCTT", "db", "dyslip", "hyper", "dp10")
# Create a subset with selected columns
df_selected <- diab[selected_cols]
head(df_selected)## # A tibble: 6 x 34
## RIDAGEYR RIAGENDR RIDRETH3 DMDEDUC2 DMDMARTL SBP DBP BMXBMI INDFMIN2
## <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 69 1 4 3 4 118 74 26.7 4
## 2 54 1 3 3 1 158 71 28.6 7
## 3 72 1 3 4 1 140 83 28.9 10
## 4 73 2 3 5 1 135 87 19.7 15
## 5 56 1 1 4 3 159 83 41.7 9
## 6 61 2 3 5 2 121 80 35.7 10
## # ... with 25 more variables: INDFMPIR <dbl>, LBXSBU <dbl>, LBXSCR <dbl>,
## # LBXSASSI <dbl>, LBXSATSI <dbl>, LBXAPB <dbl>, LBDHDD <dbl>, LBDLDL <dbl>,
## # LBXTR <dbl>, LBXTC <dbl>, LBXGLU <dbl>, LBXGLT <dbl>, ALQ101 <dbl>,
## # ALQ110 <dbl>, SMQ020 <dbl>, PAQ665 <dbl>, PAQ650 <dbl>, LBXCOT <dbl>,
## # LBXHCT <dbl>, URXCOTT <dbl>, URXHCTT <dbl>, db <dbl>, dyslip <dbl>,
## # hyper <dbl>, dp10 <dbl>5. Independent variables (Smoking, nicotine’s metabolites: cotinine, hydroxycotinine )
5.1 Cotinine and Hydroxycotinine - Serum
# LBXCOT - Cotinine, Serum (ng/mL)
# LBXHCT - Hydroxycotinine, Serum (ng/mL)5.2 Cotinine and Hydroxycotinine - Urine
# URXCOTT - Total Cotinine, urine (ng/mL)
# URXHCTT - Total Hydroxycotinine, urine (ng/mL)Checking missing
# Examing the pattern of missing variables
suppressMessages(library(rms))## Warning in .recacheSubclasses(def@className, def, env): undefined subclass
## "packedMatrix" of class "replValueSp"; definition not updated## Warning in .recacheSubclasses(def@className, def, env): undefined subclass
## "packedMatrix" of class "mMatrix"; definition not updatedna.patterns = naclus(df_selected)
plot(na.patterns)
Rename all variables in one go
df_rename <- df_selected %>% rename(age = RIDAGEYR, sex = RIAGENDR, ethnic = RIDRETH3,
educ = DMDEDUC2, masts = DMDMARTL, BMI = BMXBMI,
finc = INDFMIN2, PIR = INDFMPIR,
BUN = LBXSBU, crea = LBXSCR, AST = LBXSASSI, ALT = LBXSATSI,
apoB = LBXAPB, HDL = LBDHDD, LDL = LBDLDL,
TG = LBXTR, TC = LBXTC, Glu = LBXGLU, tolGlu = LBXGLT,
sCOT = LBXCOT, sHCOT = LBXHCT, uCOT = URXCOTT, uHCOT = URXHCTT)6. Table 1
table1(~ age + sex + ethnic + educ + masts + SBP + DBP + BMI + finc + PIR +
BUN + crea + AST + ALT + apoB + HDL + LDL + TG + TC +
Glu + tolGlu + ALQ101 + ALQ110 + SMQ020 + PAQ650 + PAQ665 +
sCOT + sHCOT + uCOT + uHCOT + db + dyslip + hyper | as.factor(dp10), data = df_rename)| 0 (N=4861) |
1 (N=511) |
Overall (N=5372) |
|
|---|---|---|---|
| Age in years at screening | |||
| Mean (SD) | 47.1 (18.6) | 50.2 (17.4) | 47.4 (18.5) |
| Median [Min, Max] | 46.0 [18.0, 80.0] | 52.0 [18.0, 80.0] | 47.0 [18.0, 80.0] |
| Gender | |||
| Mean (SD) | 1.50 (0.500) | 1.67 (0.471) | 1.52 (0.500) |
| Median [Min, Max] | 2.00 [1.00, 2.00] | 2.00 [1.00, 2.00] | 2.00 [1.00, 2.00] |
| Race/Hispanic origin w/ NH Asian | |||
| Mean (SD) | 3.29 (1.50) | 3.11 (1.43) | 3.28 (1.50) |
| Median [Min, Max] | 3.00 [1.00, 7.00] | 3.00 [1.00, 7.00] | 3.00 [1.00, 7.00] |
| Education level - Adults 20+ | |||
| Mean (SD) | 3.60 (1.19) | 3.05 (1.22) | 3.55 (1.21) |
| Median [Min, Max] | 4.00 [1.00, 9.00] | 3.00 [1.00, 5.00] | 4.00 [1.00, 9.00] |
| Missing | 297 (6.1%) | 21 (4.1%) | 318 (5.9%) |
| Marital status | |||
| Mean (SD) | 2.50 (2.14) | 2.88 (3.77) | 2.53 (2.35) |
| Median [Min, Max] | 1.00 [1.00, 77.0] | 2.50 [1.00, 77.0] | 1.00 [1.00, 77.0] |
| Missing | 297 (6.1%) | 21 (4.1%) | 318 (5.9%) |
| Systolic: Blood pres (1st rdg) mm Hg | |||
| Mean (SD) | 122 (17.1) | 125 (19.0) | 122 (17.3) |
| Median [Min, Max] | 119 [70.0, 229] | 121 [66.0, 216] | 119 [66.0, 229] |
| Missing | 430 (8.8%) | 58 (11.4%) | 488 (9.1%) |
| Diastolic: Blood pres (1st rdg) mm Hg | |||
| Mean (SD) | 68.9 (12.6) | 69.3 (12.4) | 68.9 (12.6) |
| Median [Min, Max] | 70.0 [0, 119] | 70.0 [0, 104] | 70.0 [0, 119] |
| Missing | 430 (8.8%) | 58 (11.4%) | 488 (9.1%) |
| Body Mass Index (kg/m**2) | |||
| Mean (SD) | 28.7 (6.91) | 31.6 (9.38) | 29.0 (7.23) |
| Median [Min, Max] | 27.5 [14.1, 70.1] | 30.4 [15.2, 82.9] | 27.7 [14.1, 82.9] |
| Missing | 50 (1.0%) | 6 (1.2%) | 56 (1.0%) |
| Annual family income | |||
| Mean (SD) | 10.7 (13.4) | 8.96 (14.7) | 10.6 (13.5) |
| Median [Min, Max] | 8.00 [1.00, 99.0] | 6.00 [1.00, 99.0] | 7.00 [1.00, 99.0] |
| Missing | 44 (0.9%) | 8 (1.6%) | 52 (1.0%) |
| Ratio of family income to poverty | |||
| Mean (SD) | 2.56 (1.66) | 1.76 (1.34) | 2.49 (1.65) |
| Median [Min, Max] | 2.19 [0, 5.00] | 1.27 [0, 5.00] | 2.10 [0, 5.00] |
| Missing | 344 (7.1%) | 46 (9.0%) | 390 (7.3%) |
| Blood urea nitrogen (mg/dL) | |||
| Mean (SD) | 13.2 (5.68) | 13.5 (8.99) | 13.3 (6.08) |
| Median [Min, Max] | 12.0 [1.00, 73.0] | 12.0 [2.00, 95.0] | 12.0 [1.00, 95.0] |
| Missing | 204 (4.2%) | 21 (4.1%) | 225 (4.2%) |
| Creatinine (mg/dL) | |||
| Mean (SD) | 0.906 (0.431) | 0.984 (1.06) | 0.914 (0.525) |
| Median [Min, Max] | 0.860 [0.300, 16.6] | 0.820 [0.360, 17.4] | 0.850 [0.300, 17.4] |
| Missing | 204 (4.2%) | 21 (4.1%) | 225 (4.2%) |
| Aspartate aminotransferase AST (U/L) | |||
| Mean (SD) | 25.2 (18.6) | 26.5 (21.3) | 25.4 (18.9) |
| Median [Min, Max] | 22.0 [9.00, 882] | 22.0 [11.0, 294] | 22.0 [9.00, 882] |
| Missing | 206 (4.2%) | 21 (4.1%) | 227 (4.2%) |
| Alanine aminotransferase ALT (U/L) | |||
| Mean (SD) | 24.8 (18.7) | 25.5 (21.2) | 24.9 (19.0) |
| Median [Min, Max] | 20.0 [6.00, 536] | 20.0 [7.00, 300] | 20.0 [6.00, 536] |
| Missing | 206 (4.2%) | 21 (4.1%) | 227 (4.2%) |
| Apolipoprotein (B) (mg/dL) | |||
| Mean (SD) | 88.8 (24.7) | 93.9 (26.8) | 89.2 (25.0) |
| Median [Min, Max] | 87.0 [24.0, 228] | 93.0 [20.0, 234] | 88.0 [20.0, 234] |
| Missing | 2619 (53.9%) | 284 (55.6%) | 2903 (54.0%) |
| Direct HDL-Cholesterol (mg/dL) | |||
| Mean (SD) | 52.9 (16.0) | 51.0 (15.3) | 52.7 (15.9) |
| Median [Min, Max] | 50.0 [10.0, 173] | 48.0 [22.0, 117] | 50.0 [10.0, 173] |
| Missing | 191 (3.9%) | 18 (3.5%) | 209 (3.9%) |
| LDL-cholesterol (mg/dL) | |||
| Mean (SD) | 110 (34.5) | 112 (37.2) | 110 (34.8) |
| Median [Min, Max] | 107 [14.0, 375] | 110 [24.0, 288] | 107 [14.0, 375] |
| Missing | 2652 (54.6%) | 287 (56.2%) | 2939 (54.7%) |
| Triglyceride (mg/dL) | |||
| Mean (SD) | 117 (127) | 138 (97.5) | 119 (125) |
| Median [Min, Max] | 92.0 [14.0, 4230] | 115 [21.0, 1000] | 94.0 [14.0, 4230] |
| Missing | 2618 (53.9%) | 284 (55.6%) | 2902 (54.0%) |
| Total Cholesterol( mg/dL) | |||
| Mean (SD) | 187 (41.8) | 191 (41.5) | 188 (41.8) |
| Median [Min, Max] | 184 [82.0, 813] | 187 [85.0, 380] | 184 [82.0, 813] |
| Missing | 191 (3.9%) | 18 (3.5%) | 209 (3.9%) |
| Fasting Glucose (mg/dL) | |||
| Mean (SD) | 106 (31.8) | 115 (42.9) | 107 (33.1) |
| Median [Min, Max] | 99.0 [51.0, 405] | 102 [67.0, 375] | 99.0 [51.0, 405] |
| Missing | 2608 (53.7%) | 280 (54.8%) | 2888 (53.8%) |
| Two Hour Glucose(OGTT) (mg/dL) | |||
| Mean (SD) | 117 (49.2) | 126 (51.2) | 117 (49.4) |
| Median [Min, Max] | 106 [40.0, 604] | 117 [48.0, 352] | 107 [40.0, 604] |
| Missing | 3105 (63.9%) | 347 (67.9%) | 3452 (64.3%) |
| Had at least 12 alcohol drinks/1 yr? | |||
| Mean (SD) | 1.31 (0.531) | 1.32 (0.662) | 1.31 (0.545) |
| Median [Min, Max] | 1.00 [1.00, 9.00] | 1.00 [1.00, 9.00] | 1.00 [1.00, 9.00] |
| Missing | 2 (0.0%) | 0 (0%) | 2 (0.0%) |
| Had at least 12 alcohol drinks/lifetime? | |||
| Mean (SD) | 1.59 (0.598) | 1.52 (0.501) | 1.59 (0.590) |
| Median [Min, Max] | 2.00 [1.00, 9.00] | 2.00 [1.00, 2.00] | 2.00 [1.00, 9.00] |
| Missing | 3412 (70.2%) | 361 (70.6%) | 3773 (70.2%) |
| Smoked at least 100 cigarettes in life | |||
| Mean (SD) | 1.60 (0.513) | 1.44 (0.497) | 1.58 (0.514) |
| Median [Min, Max] | 2.00 [1.00, 9.00] | 1.00 [1.00, 2.00] | 2.00 [1.00, 9.00] |
| Vigorous recreational activities | |||
| Mean (SD) | 1.75 (0.434) | 1.89 (0.310) | 1.76 (0.426) |
| Median [Min, Max] | 2.00 [1.00, 2.00] | 2.00 [1.00, 2.00] | 2.00 [1.00, 2.00] |
| Moderate recreational activities | |||
| Mean (SD) | 1.56 (0.496) | 1.74 (0.437) | 1.58 (0.494) |
| Median [Min, Max] | 2.00 [1.00, 2.00] | 2.00 [1.00, 2.00] | 2.00 [1.00, 2.00] |
| Cotinine, Serum (ng/mL) | |||
| Mean (SD) | 55.0 (129) | 100 (161) | 59.3 (133) |
| Median [Min, Max] | 0.0330 [0.0110, 1820] | 0.245 [0.0110, 1030] | 0.0370 [0.0110, 1820] |
| Missing | 189 (3.9%) | 19 (3.7%) | 208 (3.9%) |
| Hematocrit (%) | |||
| Mean (SD) | 21.8 (59.6) | 43.9 (76.7) | 23.9 (61.7) |
| Median [Min, Max] | 0.0110 [0.0110, 1150] | 0.0925 [0.0110, 540] | 0.0110 [0.0110, 1150] |
| Missing | 189 (3.9%) | 19 (3.7%) | 208 (3.9%) |
| Total Cotinine, urine (ng/mL) | |||
| Mean (SD) | 660 (1790) | 1510 (3170) | 730 (1960) |
| Median [Min, Max] | 0.430 [0.0210, 26800] | 12.8 [0.0210, 24200] | 0.486 [0.0210, 26800] |
| Missing | 3279 (67.5%) | 369 (72.2%) | 3648 (67.9%) |
| Total Hydroxycotinine, urine (ng/mL) | |||
| Mean (SD) | 1300 (4110) | 2980 (6040) | 1440 (4320) |
| Median [Min, Max] | 0.761 [0.0210, 70600] | 17.8 [0.0210, 40400] | 0.855 [0.0210, 70600] |
| Missing | 3279 (67.5%) | 369 (72.2%) | 3648 (67.9%) |
| db | |||
| Mean (SD) | 0.145 (0.352) | 0.243 (0.429) | 0.155 (0.361) |
| Median [Min, Max] | 0 [0, 1.00] | 0 [0, 1.00] | 0 [0, 1.00] |
| dyslip | |||
| Mean (SD) | 0.682 (0.466) | 0.740 (0.439) | 0.687 (0.464) |
| Median [Min, Max] | 1.00 [0, 1.00] | 1.00 [0, 1.00] | 1.00 [0, 1.00] |
| hyper | |||
| Mean (SD) | 0.340 (0.474) | 0.515 (0.500) | 0.356 (0.479) |
| Median [Min, Max] | 0 [0, 1.00] | 1.00 [0, 1.00] | 0 [0, 1.00] |
Change to factor
df_new = df_rename
df_new$dp10 = as.factor(df_new$dp10)
df_new$sex = as.factor(df_new$sex)
df_new$ethnic = as.factor(df_new$ethnic)
df_new$educ = as.factor(df_new$educ)
df_new$masts = as.factor(df_new$masts)
df_new$ALQ101 = as.factor(df_new$ALQ101)
df_new$ALQ110 = as.factor(df_new$ALQ110)
df_new$SMQ020 = as.factor(df_new$SMQ020)
df_new$PAQ650 = as.factor(df_new$PAQ650)
df_new$PAQ665 = as.factor(df_new$PAQ665)
df_new$hyper = as.factor(df_new$hyper)
df_new$db = as.factor(df_new$db)
df_new$dyslip = as.factor(df_new$dyslip)Check Table 1
tab1 = df_new
table1(~ age + sex + ethnic + educ + masts + SBP + DBP + BMI + finc + PIR +
BUN + crea + AST + ALT + apoB + HDL + LDL + TG + TC +
Glu + tolGlu + ALQ101 + ALQ110 + SMQ020 + PAQ650 + PAQ665 +
sCOT + sHCOT + uCOT + uHCOT + db + dyslip + hyper | dp10, data = tab1)| 0 (N=4861) |
1 (N=511) |
Overall (N=5372) |
|
|---|---|---|---|
| Age in years at screening | |||
| Mean (SD) | 47.1 (18.6) | 50.2 (17.4) | 47.4 (18.5) |
| Median [Min, Max] | 46.0 [18.0, 80.0] | 52.0 [18.0, 80.0] | 47.0 [18.0, 80.0] |
| sex | |||
| 1 | 2416 (49.7%) | 169 (33.1%) | 2585 (48.1%) |
| 2 | 2445 (50.3%) | 342 (66.9%) | 2787 (51.9%) |
| ethnic | |||
| 1 | 676 (13.9%) | 77 (15.1%) | 753 (14.0%) |
| 2 | 423 (8.7%) | 58 (11.4%) | 481 (9.0%) |
| 3 | 2091 (43.0%) | 224 (43.8%) | 2315 (43.1%) |
| 4 | 979 (20.1%) | 108 (21.1%) | 1087 (20.2%) |
| 6 | 544 (11.2%) | 18 (3.5%) | 562 (10.5%) |
| 7 | 148 (3.0%) | 26 (5.1%) | 174 (3.2%) |
| educ | |||
| 1 | 293 (6.0%) | 64 (12.5%) | 357 (6.6%) |
| 2 | 578 (11.9%) | 105 (20.5%) | 683 (12.7%) |
| 3 | 1012 (20.8%) | 118 (23.1%) | 1130 (21.0%) |
| 4 | 1443 (29.7%) | 148 (29.0%) | 1591 (29.6%) |
| 5 | 1237 (25.4%) | 55 (10.8%) | 1292 (24.1%) |
| 9 | 1 (0.0%) | 0 (0%) | 1 (0.0%) |
| Missing | 297 (6.1%) | 21 (4.1%) | 318 (5.9%) |
| masts | |||
| 1 | 2449 (50.4%) | 188 (36.8%) | 2637 (49.1%) |
| 2 | 312 (6.4%) | 57 (11.2%) | 369 (6.9%) |
| 3 | 484 (10.0%) | 98 (19.2%) | 582 (10.8%) |
| 4 | 124 (2.6%) | 27 (5.3%) | 151 (2.8%) |
| 5 | 872 (17.9%) | 83 (16.2%) | 955 (17.8%) |
| 6 | 322 (6.6%) | 36 (7.0%) | 358 (6.7%) |
| 77 | 1 (0.0%) | 1 (0.2%) | 2 (0.0%) |
| Missing | 297 (6.1%) | 21 (4.1%) | 318 (5.9%) |
| Systolic: Blood pres (1st rdg) mm Hg | |||
| Mean (SD) | 122 (17.1) | 125 (19.0) | 122 (17.3) |
| Median [Min, Max] | 119 [70.0, 229] | 121 [66.0, 216] | 119 [66.0, 229] |
| Missing | 430 (8.8%) | 58 (11.4%) | 488 (9.1%) |
| Diastolic: Blood pres (1st rdg) mm Hg | |||
| Mean (SD) | 68.9 (12.6) | 69.3 (12.4) | 68.9 (12.6) |
| Median [Min, Max] | 70.0 [0, 119] | 70.0 [0, 104] | 70.0 [0, 119] |
| Missing | 430 (8.8%) | 58 (11.4%) | 488 (9.1%) |
| Body Mass Index (kg/m**2) | |||
| Mean (SD) | 28.7 (6.91) | 31.6 (9.38) | 29.0 (7.23) |
| Median [Min, Max] | 27.5 [14.1, 70.1] | 30.4 [15.2, 82.9] | 27.7 [14.1, 82.9] |
| Missing | 50 (1.0%) | 6 (1.2%) | 56 (1.0%) |
| Annual family income | |||
| Mean (SD) | 10.7 (13.4) | 8.96 (14.7) | 10.6 (13.5) |
| Median [Min, Max] | 8.00 [1.00, 99.0] | 6.00 [1.00, 99.0] | 7.00 [1.00, 99.0] |
| Missing | 44 (0.9%) | 8 (1.6%) | 52 (1.0%) |
| Ratio of family income to poverty | |||
| Mean (SD) | 2.56 (1.66) | 1.76 (1.34) | 2.49 (1.65) |
| Median [Min, Max] | 2.19 [0, 5.00] | 1.27 [0, 5.00] | 2.10 [0, 5.00] |
| Missing | 344 (7.1%) | 46 (9.0%) | 390 (7.3%) |
| Blood urea nitrogen (mg/dL) | |||
| Mean (SD) | 13.2 (5.68) | 13.5 (8.99) | 13.3 (6.08) |
| Median [Min, Max] | 12.0 [1.00, 73.0] | 12.0 [2.00, 95.0] | 12.0 [1.00, 95.0] |
| Missing | 204 (4.2%) | 21 (4.1%) | 225 (4.2%) |
| Creatinine (mg/dL) | |||
| Mean (SD) | 0.906 (0.431) | 0.984 (1.06) | 0.914 (0.525) |
| Median [Min, Max] | 0.860 [0.300, 16.6] | 0.820 [0.360, 17.4] | 0.850 [0.300, 17.4] |
| Missing | 204 (4.2%) | 21 (4.1%) | 225 (4.2%) |
| Aspartate aminotransferase AST (U/L) | |||
| Mean (SD) | 25.2 (18.6) | 26.5 (21.3) | 25.4 (18.9) |
| Median [Min, Max] | 22.0 [9.00, 882] | 22.0 [11.0, 294] | 22.0 [9.00, 882] |
| Missing | 206 (4.2%) | 21 (4.1%) | 227 (4.2%) |
| Alanine aminotransferase ALT (U/L) | |||
| Mean (SD) | 24.8 (18.7) | 25.5 (21.2) | 24.9 (19.0) |
| Median [Min, Max] | 20.0 [6.00, 536] | 20.0 [7.00, 300] | 20.0 [6.00, 536] |
| Missing | 206 (4.2%) | 21 (4.1%) | 227 (4.2%) |
| Apolipoprotein (B) (mg/dL) | |||
| Mean (SD) | 88.8 (24.7) | 93.9 (26.8) | 89.2 (25.0) |
| Median [Min, Max] | 87.0 [24.0, 228] | 93.0 [20.0, 234] | 88.0 [20.0, 234] |
| Missing | 2619 (53.9%) | 284 (55.6%) | 2903 (54.0%) |
| Direct HDL-Cholesterol (mg/dL) | |||
| Mean (SD) | 52.9 (16.0) | 51.0 (15.3) | 52.7 (15.9) |
| Median [Min, Max] | 50.0 [10.0, 173] | 48.0 [22.0, 117] | 50.0 [10.0, 173] |
| Missing | 191 (3.9%) | 18 (3.5%) | 209 (3.9%) |
| LDL-cholesterol (mg/dL) | |||
| Mean (SD) | 110 (34.5) | 112 (37.2) | 110 (34.8) |
| Median [Min, Max] | 107 [14.0, 375] | 110 [24.0, 288] | 107 [14.0, 375] |
| Missing | 2652 (54.6%) | 287 (56.2%) | 2939 (54.7%) |
| Triglyceride (mg/dL) | |||
| Mean (SD) | 117 (127) | 138 (97.5) | 119 (125) |
| Median [Min, Max] | 92.0 [14.0, 4230] | 115 [21.0, 1000] | 94.0 [14.0, 4230] |
| Missing | 2618 (53.9%) | 284 (55.6%) | 2902 (54.0%) |
| Total Cholesterol( mg/dL) | |||
| Mean (SD) | 187 (41.8) | 191 (41.5) | 188 (41.8) |
| Median [Min, Max] | 184 [82.0, 813] | 187 [85.0, 380] | 184 [82.0, 813] |
| Missing | 191 (3.9%) | 18 (3.5%) | 209 (3.9%) |
| Fasting Glucose (mg/dL) | |||
| Mean (SD) | 106 (31.8) | 115 (42.9) | 107 (33.1) |
| Median [Min, Max] | 99.0 [51.0, 405] | 102 [67.0, 375] | 99.0 [51.0, 405] |
| Missing | 2608 (53.7%) | 280 (54.8%) | 2888 (53.8%) |
| Two Hour Glucose(OGTT) (mg/dL) | |||
| Mean (SD) | 117 (49.2) | 126 (51.2) | 117 (49.4) |
| Median [Min, Max] | 106 [40.0, 604] | 117 [48.0, 352] | 107 [40.0, 604] |
| Missing | 3105 (63.9%) | 347 (67.9%) | 3452 (64.3%) |
| ALQ101 | |||
| 1 | 3410 (70.2%) | 361 (70.6%) | 3771 (70.2%) |
| 2 | 1443 (29.7%) | 148 (29.0%) | 1591 (29.6%) |
| 9 | 6 (0.1%) | 2 (0.4%) | 8 (0.1%) |
| Missing | 2 (0.0%) | 0 (0%) | 2 (0.0%) |
| ALQ110 | |||
| 1 | 610 (12.5%) | 72 (14.1%) | 682 (12.7%) |
| 2 | 836 (17.2%) | 78 (15.3%) | 914 (17.0%) |
| 9 | 3 (0.1%) | 0 (0%) | 3 (0.1%) |
| Missing | 3412 (70.2%) | 361 (70.6%) | 3773 (70.2%) |
| SMQ020 | |||
| 1 | 1966 (40.4%) | 287 (56.2%) | 2253 (41.9%) |
| 2 | 2893 (59.5%) | 224 (43.8%) | 3117 (58.0%) |
| 9 | 2 (0.0%) | 0 (0%) | 2 (0.0%) |
| PAQ650 | |||
| 1 | 1225 (25.2%) | 55 (10.8%) | 1280 (23.8%) |
| 2 | 3636 (74.8%) | 456 (89.2%) | 4092 (76.2%) |
| PAQ665 | |||
| 1 | 2129 (43.8%) | 131 (25.6%) | 2260 (42.1%) |
| 2 | 2732 (56.2%) | 380 (74.4%) | 3112 (57.9%) |
| Cotinine, Serum (ng/mL) | |||
| Mean (SD) | 55.0 (129) | 100 (161) | 59.3 (133) |
| Median [Min, Max] | 0.0330 [0.0110, 1820] | 0.245 [0.0110, 1030] | 0.0370 [0.0110, 1820] |
| Missing | 189 (3.9%) | 19 (3.7%) | 208 (3.9%) |
| Hematocrit (%) | |||
| Mean (SD) | 21.8 (59.6) | 43.9 (76.7) | 23.9 (61.7) |
| Median [Min, Max] | 0.0110 [0.0110, 1150] | 0.0925 [0.0110, 540] | 0.0110 [0.0110, 1150] |
| Missing | 189 (3.9%) | 19 (3.7%) | 208 (3.9%) |
| Total Cotinine, urine (ng/mL) | |||
| Mean (SD) | 660 (1790) | 1510 (3170) | 730 (1960) |
| Median [Min, Max] | 0.430 [0.0210, 26800] | 12.8 [0.0210, 24200] | 0.486 [0.0210, 26800] |
| Missing | 3279 (67.5%) | 369 (72.2%) | 3648 (67.9%) |
| Total Hydroxycotinine, urine (ng/mL) | |||
| Mean (SD) | 1300 (4110) | 2980 (6040) | 1440 (4320) |
| Median [Min, Max] | 0.761 [0.0210, 70600] | 17.8 [0.0210, 40400] | 0.855 [0.0210, 70600] |
| Missing | 3279 (67.5%) | 369 (72.2%) | 3648 (67.9%) |
| db | |||
| 0 | 4155 (85.5%) | 387 (75.7%) | 4542 (84.5%) |
| 1 | 706 (14.5%) | 124 (24.3%) | 830 (15.5%) |
| dyslip | |||
| 0 | 1548 (31.8%) | 133 (26.0%) | 1681 (31.3%) |
| 1 | 3313 (68.2%) | 378 (74.0%) | 3691 (68.7%) |
| hyper | |||
| 0 | 3210 (66.0%) | 248 (48.5%) | 3458 (64.4%) |
| 1 | 1651 (34.0%) | 263 (51.5%) | 1914 (35.6%) |
7. Descriptive analysis
comp = compareGroups(dp10 ~ age + sex + ethnic + educ + masts + SBP + DBP + BMI + finc + PIR +
BUN + crea + AST + ALT + apoB + HDL + LDL + TG + TC +
Glu + tolGlu + ALQ101 + ALQ110 + SMQ020 + PAQ650 + PAQ665 +
sCOT + sHCOT + uCOT + uHCOT + db + dyslip + hyper , data = tab1)## Warning in chisq.test(xx, correct = FALSE): Chi-squared approximation may be
## incorrect
## Warning in chisq.test(xx, correct = FALSE): Chi-squared approximation may be
## incorrectcreateTable(comp)##
## --------Summary descriptives table by 'dp10'---------
##
## ________________________________________________________________________
## 0 1 p.overall
## N=4861 N=511
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
## Age in years at screening 47.1 (18.6) 50.2 (17.4) <0.001
## sex: <0.001
## 1 2416 (49.7%) 169 (33.1%)
## 2 2445 (50.3%) 342 (66.9%)
## ethnic: <0.001
## 1 676 (13.9%) 77 (15.1%)
## 2 423 (8.70%) 58 (11.4%)
## 3 2091 (43.0%) 224 (43.8%)
## 4 979 (20.1%) 108 (21.1%)
## 6 544 (11.2%) 18 (3.52%)
## 7 148 (3.04%) 26 (5.09%)
## educ: .
## 1 293 (6.42%) 64 (13.1%)
## 2 578 (12.7%) 105 (21.4%)
## 3 1012 (22.2%) 118 (24.1%)
## 4 1443 (31.6%) 148 (30.2%)
## 5 1237 (27.1%) 55 (11.2%)
## 9 1 (0.02%) 0 (0.00%)
## masts: .
## 1 2449 (53.7%) 188 (38.4%)
## 2 312 (6.84%) 57 (11.6%)
## 3 484 (10.6%) 98 (20.0%)
## 4 124 (2.72%) 27 (5.51%)
## 5 872 (19.1%) 83 (16.9%)
## 6 322 (7.06%) 36 (7.35%)
## 77 1 (0.02%) 1 (0.20%)
## Systolic: Blood pres (1st rdg) mm Hg 122 (17.1) 125 (19.0) 0.001
## Diastolic: Blood pres (1st rdg) mm Hg 68.9 (12.6) 69.3 (12.4) 0.538
## Body Mass Index (kg/m**2) 28.7 (6.91) 31.6 (9.38) <0.001
## Annual family income 10.7 (13.4) 8.96 (14.7) 0.010
## Ratio of family income to poverty 2.56 (1.66) 1.76 (1.34) <0.001
## Blood urea nitrogen (mg/dL) 13.2 (5.68) 13.5 (8.99) 0.543
## Creatinine (mg/dL) 0.91 (0.43) 0.98 (1.06) 0.108
## Aspartate aminotransferase AST (U/L) 25.2 (18.6) 26.5 (21.3) 0.208
## Alanine aminotransferase ALT (U/L) 24.8 (18.7) 25.5 (21.2) 0.488
## Apolipoprotein (B) (mg/dL) 88.8 (24.7) 93.9 (26.8) 0.006
## Direct HDL-Cholesterol (mg/dL) 52.9 (16.0) 51.0 (15.3) 0.010
## LDL-cholesterol (mg/dL) 110 (34.5) 112 (37.2) 0.397
## Triglyceride (mg/dL) 117 (127) 138 (97.5) 0.003
## Total Cholesterol( mg/dL) 187 (41.8) 191 (41.5) 0.107
## Fasting Glucose (mg/dL) 106 (31.8) 115 (42.9) 0.001
## Two Hour Glucose(OGTT) (mg/dL) 117 (49.2) 126 (51.2) 0.026
## ALQ101: 0.260
## 1 3410 (70.2%) 361 (70.6%)
## 2 1443 (29.7%) 148 (29.0%)
## 9 6 (0.12%) 2 (0.39%)
## ALQ110: 0.400
## 1 610 (42.1%) 72 (48.0%)
## 2 836 (57.7%) 78 (52.0%)
## 9 3 (0.21%) 0 (0.00%)
## SMQ020: <0.001
## 1 1966 (40.4%) 287 (56.2%)
## 2 2893 (59.5%) 224 (43.8%)
## 9 2 (0.04%) 0 (0.00%)
## PAQ650: <0.001
## 1 1225 (25.2%) 55 (10.8%)
## 2 3636 (74.8%) 456 (89.2%)
## PAQ665: <0.001
## 1 2129 (43.8%) 131 (25.6%)
## 2 2732 (56.2%) 380 (74.4%)
## Cotinine, Serum (ng/mL) 55.0 (129) 100 (161) <0.001
## Hematocrit (%) 21.8 (59.6) 43.9 (76.7) <0.001
## Total Cotinine, urine (ng/mL) 660 (1792) 1511 (3172) 0.002
## Total Hydroxycotinine, urine (ng/mL) 1301 (4108) 2975 (6038) 0.001
## db: <0.001
## 0 4155 (85.5%) 387 (75.7%)
## 1 706 (14.5%) 124 (24.3%)
## dyslip: 0.008
## 0 1548 (31.8%) 133 (26.0%)
## 1 3313 (68.2%) 378 (74.0%)
## hyper: <0.001
## 0 3210 (66.0%) 248 (48.5%)
## 1 1651 (34.0%) 263 (51.5%)
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯- Or
tab = descrTable(dp10 ~ age + sex + ethnic + educ + masts + SBP + DBP + BMI + finc + PIR +
BUN + crea + AST + ALT + apoB + HDL + LDL + TG + TC +
Glu + tolGlu + ALQ101 + ALQ110 + SMQ020 + PAQ650 + PAQ665 +
sCOT + sHCOT + uCOT + uHCOT + db + dyslip + hyper , data = tab1)## Warning in chisq.test(xx, correct = FALSE): Chi-squared approximation may be
## incorrect
## Warning in chisq.test(xx, correct = FALSE): Chi-squared approximation may be
## incorrect# Example of case-control study: Odds Ratios
descrTable(dp10 ~ ., tab1, show.ratio=TRUE, show.p.overall=FALSE)## Warning in chisq.test(xx, correct = FALSE): Chi-squared approximation may be
## incorrect
## Warning in chisq.test(xx, correct = FALSE): Chi-squared approximation may be
## incorrect##
## --------Summary descriptives table by 'dp10'---------
##
## _______________________________________________________________________________________
## 0 1 OR p.ratio
## N=4861 N=511
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
## Age in years at screening 47.1 (18.6) 50.2 (17.4) 1.01 [1.00;1.01] <0.001
## sex:
## 1 2416 (49.7%) 169 (33.1%) Ref. Ref.
## 2 2445 (50.3%) 342 (66.9%) 2.00 [1.65;2.43] <0.001
## ethnic:
## 1 676 (13.9%) 77 (15.1%) Ref. Ref.
## 2 423 (8.70%) 58 (11.4%) 1.20 [0.84;1.73] 0.317
## 3 2091 (43.0%) 224 (43.8%) 0.94 [0.72;1.24] 0.655
## 4 979 (20.1%) 108 (21.1%) 0.97 [0.71;1.32] 0.836
## 6 544 (11.2%) 18 (3.52%) 0.29 [0.17;0.48] <0.001
## 7 148 (3.04%) 26 (5.09%) 1.55 [0.94;2.47] 0.083
## educ:
## 1 293 (6.42%) 64 (13.1%) Ref. Ref.
## 2 578 (12.7%) 105 (21.4%) . [.;.] .
## 3 1012 (22.2%) 118 (24.1%) . [.;.] .
## 4 1443 (31.6%) 148 (30.2%) . [.;.] .
## 5 1237 (27.1%) 55 (11.2%) . [.;.] .
## 9 1 (0.02%) 0 (0.00%) . [.;.] .
## masts:
## 1 2449 (53.7%) 188 (38.4%) Ref. Ref.
## 2 312 (6.84%) 57 (11.6%) 2.38 [1.72;3.26] <0.001
## 3 484 (10.6%) 98 (20.0%) 2.64 [2.02;3.42] <0.001
## 4 124 (2.72%) 27 (5.51%) 2.85 [1.80;4.37] <0.001
## 5 872 (19.1%) 83 (16.9%) 1.24 [0.94;1.62] 0.121
## 6 322 (7.06%) 36 (7.35%) 1.46 [0.99;2.10] 0.056
## 77 1 (0.02%) 1 (0.20%) 13.0 [0.33;506] 0.143
## Systolic: Blood pres (1st rdg) mm Hg 122 (17.1) 125 (19.0) 1.01 [1.00;1.01] <0.001
## Diastolic: Blood pres (1st rdg) mm Hg 68.9 (12.6) 69.3 (12.4) 1.00 [0.99;1.01] 0.542
## Body Mass Index (kg/m**2) 28.7 (6.91) 31.6 (9.38) 1.05 [1.04;1.06] <0.001
## Annual family income 10.7 (13.4) 8.96 (14.7) 0.99 [0.98;1.00] 0.007
## Ratio of family income to poverty 2.56 (1.66) 1.76 (1.34) 0.71 [0.67;0.76] <0.001
## Blood urea nitrogen (mg/dL) 13.2 (5.68) 13.5 (8.99) 1.01 [0.99;1.02] 0.382
## Creatinine (mg/dL) 0.91 (0.43) 0.98 (1.06) 1.18 [1.05;1.34] 0.007
## Aspartate aminotransferase AST (U/L) 25.2 (18.6) 26.5 (21.3) 1.00 [1.00;1.01] 0.189
## Alanine aminotransferase ALT (U/L) 24.8 (18.7) 25.5 (21.2) 1.00 [1.00;1.01] 0.445
## Apolipoprotein (B) (mg/dL) 88.8 (24.7) 93.9 (26.8) 1.01 [1.00;1.01] 0.003
## Direct HDL-Cholesterol (mg/dL) 52.9 (16.0) 51.0 (15.3) 0.99 [0.99;1.00] 0.012
## LDL-cholesterol (mg/dL) 110 (34.5) 112 (37.2) 1.00 [1.00;1.01] 0.366
## Triglyceride (mg/dL) 117 (127) 138 (97.5) 1.00 [1.00;1.00] 0.052
## Total Cholesterol( mg/dL) 187 (41.8) 191 (41.5) 1.00 [1.00;1.00] 0.108
## Fasting Glucose (mg/dL) 106 (31.8) 115 (42.9) 1.01 [1.00;1.01] <0.001
## Two Hour Glucose(OGTT) (mg/dL) 117 (49.2) 126 (51.2) 1.00 [1.00;1.01] 0.022
## ALQ101:
## 1 3410 (70.2%) 361 (70.6%) Ref. Ref.
## 2 1443 (29.7%) 148 (29.0%) 0.97 [0.79;1.18] 0.762
## 9 6 (0.12%) 2 (0.39%) 3.30 [0.44;14.9] 0.209
## ALQ110:
## 1 610 (42.1%) 72 (48.0%) Ref. Ref.
## 2 836 (57.7%) 78 (52.0%) . [.;.] .
## 9 3 (0.21%) 0 (0.00%) . [.;.] .
## SMQ020:
## 1 1966 (40.4%) 287 (56.2%) Ref. Ref.
## 2 2893 (59.5%) 224 (43.8%) . [.;.] .
## 9 2 (0.04%) 0 (0.00%) . [.;.] .
## PAQ665:
## 1 2129 (43.8%) 131 (25.6%) Ref. Ref.
## 2 2732 (56.2%) 380 (74.4%) 2.26 [1.84;2.79] <0.001
## PAQ650:
## 1 1225 (25.2%) 55 (10.8%) Ref. Ref.
## 2 3636 (74.8%) 456 (89.2%) 2.79 [2.11;3.75] <0.001
## Cotinine, Serum (ng/mL) 55.0 (129) 100 (161) 1.00 [1.00;1.00] <0.001
## Hematocrit (%) 21.8 (59.6) 43.9 (76.7) 1.00 [1.00;1.01] <0.001
## Total Cotinine, urine (ng/mL) 660 (1792) 1511 (3172) 1.00 [1.00;1.00] <0.001
## Total Hydroxycotinine, urine (ng/mL) 1301 (4108) 2975 (6038) 1.00 [1.00;1.00] <0.001
## db:
## 0 4155 (85.5%) 387 (75.7%) Ref. Ref.
## 1 706 (14.5%) 124 (24.3%) 1.89 [1.51;2.34] <0.001
## dyslip:
## 0 1548 (31.8%) 133 (26.0%) Ref. Ref.
## 1 3313 (68.2%) 378 (74.0%) 1.33 [1.08;1.64] 0.006
## hyper:
## 0 3210 (66.0%) 248 (48.5%) Ref. Ref.
## 1 1651 (34.0%) 263 (51.5%) 2.06 [1.72;2.48] <0.001
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯- Plot
plot(tab["sex"]) # barplot
plot(tab["ethnic"])
plot(tab["educ"])
plot(tab["masts"])
plot(tab["ALQ101"])
plot(tab["ALQ110"])
plot(tab["SMQ020"])
plot(tab["PAQ650"])
plot(tab["PAQ665"])
plot(tab["db"])
plot(tab["dyslip"])
plot(tab["hyper"])
plot(tab["age"]) # histogram and normality plot## Warning in norm.plot(x = x.var, file = file.i, var.label.x = var.labels[i], :
## p-value for normality in Age in years at screening could not be calculated
plot(tab["SBP"])
plot(tab["DBP"])
plot(tab["finc"])## Warning in norm.plot(x = x.var, file = file.i, var.label.x = var.labels[i], :
## p-value for normality in Annual family income could not be calculated
plot(tab["PIR"])
plot(tab["BUN"])## Warning in norm.plot(x = x.var, file = file.i, var.label.x = var.labels[i], :
## p-value for normality in Blood urea nitrogen (mg/dL) could not be calculated
plot(tab["crea"])## Warning in norm.plot(x = x.var, file = file.i, var.label.x = var.labels[i], :
## p-value for normality in Creatinine (mg/dL) could not be calculated
plot(tab["apoB"])
plot(tab["TC"])## Warning in norm.plot(x = x.var, file = file.i, var.label.x = var.labels[i], :
## p-value for normality in Total Cholesterol( mg/dL) could not be calculated
plot(tab["TG"])
plot(tab["LDL"])
plot(tab["HDL"])## Warning in norm.plot(x = x.var, file = file.i, var.label.x = var.labels[i], :
## p-value for normality in Direct HDL-Cholesterol (mg/dL) could not be calculated
plot(tab["Glu"])
plot(tab["tolGlu"])
plot(tab["sCOT"])## Warning in norm.plot(x = x.var, file = file.i, var.label.x = var.labels[i], :
## p-value for normality in Cotinine, Serum (ng/mL) could not be calculated
plot(tab["sHCOT"])## Warning in norm.plot(x = x.var, file = file.i, var.label.x = var.labels[i], :
## p-value for normality in Hematocrit (%) could not be calculated
plot(tab["uCOT"])
plot(tab["uHCOT"])
# export2xls(tab, file = "example.xlsx")
export2word(tab, file = "tab_result.docx")8. Logistic Regression
8.1 Unadjusted model
unadj1 = glm(dp10 ~ sCOT, family = binomial, data = tab1)
unadj2 = glm(dp10 ~ sHCOT, family = binomial, data = tab1)
unadj3 = glm(dp10 ~ uCOT, family = binomial, data = tab1)
unadj4 = glm(dp10 ~ uHCOT, family = binomial, data = tab1)# Summary the results of logistic regression
summary(unadj1)##
## Call:
## glm(formula = dp10 ~ sCOT, family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.6936 -0.4177 -0.4174 -0.4173 2.2289
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.3969586 0.0541640 -44.254 < 2e-16 ***
## sCOT 0.0019554 0.0002839 6.889 5.62e-12 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 3248.9 on 5163 degrees of freedom
## Residual deviance: 3206.4 on 5162 degrees of freedom
## (208 observations deleted due to missingness)
## AIC: 3210.4
##
## Number of Fisher Scoring iterations: 5logistic.display(unadj1)##
## Logistic regression predicting dp10 : 1 vs 0
##
## OR(95%CI) P(Wald's test) P(LR-test)
## sCOT (cont. var.) 1.002 (1.0014,1.0025) < 0.001 < 0.001
##
## Log-likelihood = -1603.1901
## No. of observations = 5164
## AIC value = 3210.3801summary(unadj2)##
## Call:
## glm(formula = dp10 ~ sHCOT, family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.1650 -0.4223 -0.4220 -0.4220 2.2195
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.3739862 0.0524241 -45.284 < 2e-16 ***
## sHCOT 0.0040145 0.0005863 6.847 7.53e-12 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 3248.9 on 5163 degrees of freedom
## Residual deviance: 3205.9 on 5162 degrees of freedom
## (208 observations deleted due to missingness)
## AIC: 3209.9
##
## Number of Fisher Scoring iterations: 5logistic.display(unadj2)##
## Logistic regression predicting dp10 : 1 vs 0
##
## OR(95%CI) P(Wald's test) P(LR-test)
## sHCOT (cont. var.) 1.004 (1.0029,1.0052) < 0.001 < 0.001
##
## Log-likelihood = -1602.9731
## No. of observations = 5164
## AIC value = 3209.9462summary(unadj3)##
## Call:
## glm(formula = dp10 ~ uCOT, family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.7030 -0.3887 -0.3882 -0.3882 2.2904
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.548e+00 9.665e-02 -26.359 < 2e-16 ***
## uCOT 1.392e-04 3.225e-05 4.316 1.59e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 981.00 on 1723 degrees of freedom
## Residual deviance: 963.89 on 1722 degrees of freedom
## (3648 observations deleted due to missingness)
## AIC: 967.89
##
## Number of Fisher Scoring iterations: 5logistic.display(unadj3)##
## Logistic regression predicting dp10 : 1 vs 0
##
## OR(95%CI) P(Wald's test) P(LR-test)
## uCOT (cont. var.) 1.0001 (1.0001,1.0002) < 0.001 < 0.001
##
## Log-likelihood = -481.9445
## No. of observations = 1724
## AIC value = 967.889summary(unadj4)##
## Call:
## glm(formula = dp10 ~ uHCOT, family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.7560 -0.3948 -0.3944 -0.3944 2.2770
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.515e+00 9.431e-02 -26.662 < 2e-16 ***
## uHCOT 5.405e-05 1.416e-05 3.816 0.000135 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 981.00 on 1723 degrees of freedom
## Residual deviance: 967.82 on 1722 degrees of freedom
## (3648 observations deleted due to missingness)
## AIC: 971.82
##
## Number of Fisher Scoring iterations: 5logistic.display(unadj4)##
## Logistic regression predicting dp10 : 1 vs 0
##
## OR(95%CI) P(Wald's test) P(LR-test)
## uHCOT (cont. var.) 1.0001 (1,1.0001) < 0.001 < 0.001
##
## Log-likelihood = -483.9118
## No. of observations = 1724
## AIC value = 971.8236ggcoef(unadj1, exponentiate=T, exclude_intercept=T, vline_color = "red",
errorbar_color = "blue", errorbar_height = 0.10)
ggcoef(unadj2, exponentiate=T, exclude_intercept=T, vline_color = "red",
errorbar_color = "blue", errorbar_height = 0.10)
ggcoef(unadj3, exponentiate=T, exclude_intercept=T, vline_color = "red",
errorbar_color = "blue", errorbar_height = 0.10)
ggcoef(unadj4, exponentiate=T, exclude_intercept=T, vline_color = "red",
errorbar_color = "blue", errorbar_height = 0.10)
8.2 Model 1
### Adjust for age, sex
Model1_1 = glm(dp10 ~ sCOT + age + sex, family = binomial, data = tab1)
Model1_2 = glm(dp10 ~ sHCOT + age + sex, family = binomial, data = tab1)
Model1_3 = glm(dp10 ~ uCOT + age + sex, family = binomial, data = tab1)
Model1_4 = glm(dp10 ~ uHCOT + age + sex, family = binomial, data = tab1)
# Summary the results of logistic regression
summary(Model1_1)##
## Call:
## glm(formula = dp10 ~ sCOT + age + sex, family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.8971 -0.4894 -0.4261 -0.3237 2.5275
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -3.3075971 0.1607865 -20.571 < 2e-16 ***
## sCOT 0.0023307 0.0002925 7.968 1.61e-15 ***
## age 0.0085565 0.0026608 3.216 0.0013 **
## sex2 0.7915051 0.1025770 7.716 1.20e-14 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 3248.9 on 5163 degrees of freedom
## Residual deviance: 3132.2 on 5160 degrees of freedom
## (208 observations deleted due to missingness)
## AIC: 3140.2
##
## Number of Fisher Scoring iterations: 5logistic.display(Model1_1)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI) P(Wald's test)
## sCOT (cont. var.) 1.002 (1.0014,1.0025) 1.0023 (1.0018,1.0029) < 0.001
##
## age (cont. var.) 1.0079 (1.0028,1.013) 1.0086 (1.0033,1.0139) 0.001
##
## sex: 2 vs 1 2.01 (1.65,2.44) 2.21 (1.8,2.7) < 0.001
##
## P(LR-test)
## sCOT (cont. var.) < 0.001
##
## age (cont. var.) 0.001
##
## sex: 2 vs 1 < 0.001
##
## Log-likelihood = -1566.0897
## No. of observations = 5164
## AIC value = 3140.1794summary(Model1_2)##
## Call:
## glm(formula = dp10 ~ sHCOT + age + sex, family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.2132 -0.4900 -0.4182 -0.3346 2.4864
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -3.178005 0.156245 -20.340 < 2e-16 ***
## sHCOT 0.004303 0.000592 7.269 3.62e-13 ***
## age 0.007350 0.002648 2.775 0.00552 **
## sex2 0.744043 0.101574 7.325 2.39e-13 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 3248.9 on 5163 degrees of freedom
## Residual deviance: 3140.7 on 5160 degrees of freedom
## (208 observations deleted due to missingness)
## AIC: 3148.7
##
## Number of Fisher Scoring iterations: 5logistic.display(Model1_2)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## sHCOT (cont. var.) 1.004 (1.0029,1.0052) 1.0043 (1.0031,1.0055)
##
## age (cont. var.) 1.0079 (1.0028,1.013) 1.0074 (1.0022,1.0126)
##
## sex: 2 vs 1 2.01 (1.65,2.44) 2.1 (1.72,2.57)
##
## P(Wald's test) P(LR-test)
## sHCOT (cont. var.) < 0.001 < 0.001
##
## age (cont. var.) 0.006 0.005
##
## sex: 2 vs 1 < 0.001 < 0.001
##
## Log-likelihood = -1570.3586
## No. of observations = 5164
## AIC value = 3148.7172summary(Model1_3)##
## Call:
## glm(formula = dp10 ~ uCOT + age + sex, family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.7946 -0.4542 -0.3986 -0.3109 2.5367
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -3.286e+00 2.905e-01 -11.313 < 2e-16 ***
## uCOT 1.650e-04 3.359e-05 4.913 8.98e-07 ***
## age 5.968e-03 4.961e-03 1.203 0.229
## sex2 7.504e-01 1.883e-01 3.984 6.78e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 981.00 on 1723 degrees of freedom
## Residual deviance: 945.75 on 1720 degrees of freedom
## (3648 observations deleted due to missingness)
## AIC: 953.75
##
## Number of Fisher Scoring iterations: 5logistic.display(Model1_3)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## uCOT (cont. var.) 1.0001 (1.0001,1.0002) 1.0002 (1.0001,1.0002)
##
## age (cont. var.) 1.0039 (0.9945,1.0134) 1.006 (0.9963,1.0158)
##
## sex: 2 vs 1 1.87 (1.31,2.67) 2.12 (1.46,3.06)
##
## P(Wald's test) P(LR-test)
## uCOT (cont. var.) < 0.001 < 0.001
##
## age (cont. var.) 0.229 0.229
##
## sex: 2 vs 1 < 0.001 < 0.001
##
## Log-likelihood = -472.8738
## No. of observations = 1724
## AIC value = 953.7476summary(Model1_4)##
## Call:
## glm(formula = dp10 ~ uHCOT + age + sex, family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.7924 -0.4554 -0.3817 -0.3201 2.4964
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -3.162e+00 2.820e-01 -11.213 < 2e-16 ***
## uHCOT 6.151e-05 1.440e-05 4.271 1.94e-05 ***
## age 4.806e-03 4.891e-03 0.983 0.325780
## sex2 7.074e-01 1.864e-01 3.794 0.000148 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 981.00 on 1723 degrees of freedom
## Residual deviance: 951.74 on 1720 degrees of freedom
## (3648 observations deleted due to missingness)
## AIC: 959.74
##
## Number of Fisher Scoring iterations: 5logistic.display(Model1_4)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## uHCOT (cont. var.) 1.0001 (1,1.0001) 1.0001 (1,1.0001)
##
## age (cont. var.) 1.0039 (0.9945,1.0134) 1.0048 (0.9952,1.0145)
##
## sex: 2 vs 1 1.87 (1.31,2.67) 2.03 (1.41,2.92)
##
## P(Wald's test) P(LR-test)
## uHCOT (cont. var.) < 0.001 < 0.001
##
## age (cont. var.) 0.326 0.326
##
## sex: 2 vs 1 < 0.001 < 0.001
##
## Log-likelihood = -475.8704
## No. of observations = 1724
## AIC value = 959.7408# Plot forest
forest_Model1_1 = ggcoef(Model1_1, exponentiate=T, exclude_intercept=T, vline_color = "red",
errorbar_color = "blue", errorbar_height = 0.10)
forest_Model1_1
forest_Model1_2 = ggcoef(Model1_2, exponentiate=T, exclude_intercept=T, vline_color = "red",
errorbar_color = "blue", errorbar_height = 0.10)
forest_Model1_2
forest_Model1_3 = ggcoef(Model1_3, exponentiate=T, exclude_intercept=T, vline_color = "red",
errorbar_color = "blue", errorbar_height = 0.10)
forest_Model1_3
forest_Model1_4 = ggcoef(Model1_4, exponentiate=T, exclude_intercept=T, vline_color = "red",
errorbar_color = "blue", errorbar_height = 0.10)
forest_Model1_4
## 8.3 Model 2
### Adjust for: age, sex, ethnic group, educ, marrital status
Model2_1 = glm(dp10 ~ sCOT + age + sex + ethnic + educ + masts, family = binomial, data = tab1)
Model2_2 = glm(dp10 ~ sHCOT + age + sex + ethnic + educ + masts, family = binomial, data = tab1)
Model2_3 = glm(dp10 ~ uCOT + age + sex + ethnic + educ + masts, family = binomial, data = tab1)
Model2_4 = glm(dp10 ~ uHCOT + age + sex + ethnic + educ + masts, family = binomial, data = tab1)
# Summary the results of logistic regression
summary(Model2_1)##
## Call:
## glm(formula = dp10 ~ sCOT + age + sex + ethnic + educ + masts,
## family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.6250 -0.4800 -0.3730 -0.2796 3.0172
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.604494 0.276464 -9.421 < 2e-16 ***
## sCOT 0.001714 0.000317 5.407 6.41e-08 ***
## age 0.004208 0.003611 1.165 0.24389
## sex2 0.753555 0.108460 6.948 3.71e-12 ***
## ethnic2 0.299596 0.202146 1.482 0.13832
## ethnic3 0.158154 0.170276 0.929 0.35299
## ethnic4 0.081527 0.186236 0.438 0.66156
## ethnic6 -0.711674 0.296392 -2.401 0.01634 *
## ethnic7 0.703979 0.279101 2.522 0.01166 *
## educ2 -0.285626 0.192238 -1.486 0.13733
## educ3 -0.747469 0.192145 -3.890 0.00010 ***
## educ4 -0.828825 0.187870 -4.412 1.03e-05 ***
## educ5 -1.425360 0.222631 -6.402 1.53e-10 ***
## educ9 -10.703958 324.743793 -0.033 0.97371
## masts2 0.370365 0.187730 1.973 0.04851 *
## masts3 0.786668 0.142393 5.525 3.30e-08 ***
## masts4 0.631420 0.239130 2.640 0.00828 **
## masts5 0.193217 0.157229 1.229 0.21911
## masts6 0.188095 0.204951 0.918 0.35875
## masts77 2.934699 1.431036 2.051 0.04029 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 3101.7 on 4878 degrees of freedom
## Residual deviance: 2853.8 on 4859 degrees of freedom
## (493 observations deleted due to missingness)
## AIC: 2893.8
##
## Number of Fisher Scoring iterations: 11logistic.display(Model2_1)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## sCOT (cont. var.) 1.002 (1.0014,1.0025) 1.0017 (1.0011,1.0023)
##
## age (cont. var.) 1.0074 (1.002,1.0129) 1.0042 (0.9971,1.0113)
##
## sex: 2 vs 1 2.07 (1.69,2.54) 2.12 (1.72,2.63)
##
## ethnic: ref.=1
## 2 1.29 (0.88,1.88) 1.35 (0.91,2.01)
## 3 0.96 (0.72,1.29) 1.17 (0.84,1.64)
## 4 1.06 (0.77,1.47) 1.08 (0.75,1.56)
## 6 0.29 (0.17,0.51) 0.49 (0.27,0.88)
## 7 1.57 (0.94,2.62) 2.02 (1.17,3.49)
##
## educ: ref.=1
## 2 0.85 (0.6,1.21) 0.75 (0.52,1.1)
## 3 0.54 (0.38,0.76) 0.47 (0.32,0.69)
## 4 0.49 (0.36,0.68) 0.44 (0.3,0.63)
## 5 0.21 (0.14,0.31) 0.24 (0.16,0.37)
## 9 0 (0,4.35136067215608e+271) 0 (0,5.93946849867773e+271)
##
## masts: ref.=1
## 2 2.3 (1.66,3.19) 1.45 (1,2.09)
## 3 2.67 (2.04,3.49) 2.2 (1.66,2.9)
## 4 2.95 (1.88,4.63) 1.88 (1.18,3)
## 5 1.26 (0.96,1.66) 1.21 (0.89,1.65)
## 6 1.55 (1.06,2.25) 1.21 (0.81,1.8)
## 77 13.14 (0.82,211) 18.82 (1.14,310.9)
##
## P(Wald's test) P(LR-test)
## sCOT (cont. var.) < 0.001 < 0.001
##
## age (cont. var.) 0.244 0.244
##
## sex: 2 vs 1 < 0.001 < 0.001
##
## ethnic: ref.=1 < 0.001
## 2 0.138
## 3 0.353
## 4 0.662
## 6 0.016
## 7 0.012
##
## educ: ref.=1 < 0.001
## 2 0.137
## 3 < 0.001
## 4 < 0.001
## 5 < 0.001
## 9 0.974
##
## masts: ref.=1 < 0.001
## 2 0.049
## 3 < 0.001
## 4 0.008
## 5 0.219
## 6 0.359
## 77 0.04
##
## Log-likelihood = -1426.9079
## No. of observations = 4879
## AIC value = 2893.8158summary(Model2_2)##
## Call:
## glm(formula = dp10 ~ sHCOT + age + sex + ethnic + educ + masts,
## family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.8117 -0.4797 -0.3681 -0.2821 3.0076
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.533e+00 2.747e-01 -9.223 < 2e-16 ***
## sHCOT 3.055e-03 6.241e-04 4.895 9.84e-07 ***
## age 3.290e-03 3.596e-03 0.915 0.360217
## sex2 7.195e-01 1.075e-01 6.692 2.21e-11 ***
## ethnic2 3.125e-01 2.018e-01 1.549 0.121420
## ethnic3 1.832e-01 1.698e-01 1.079 0.280592
## ethnic4 1.365e-01 1.844e-01 0.740 0.459195
## ethnic6 -6.832e-01 2.960e-01 -2.308 0.020995 *
## ethnic7 7.482e-01 2.780e-01 2.691 0.007117 **
## educ2 -2.638e-01 1.921e-01 -1.373 0.169702
## educ3 -7.375e-01 1.922e-01 -3.837 0.000125 ***
## educ4 -8.340e-01 1.880e-01 -4.437 9.10e-06 ***
## educ5 -1.448e+00 2.225e-01 -6.508 7.62e-11 ***
## educ9 -1.075e+01 3.247e+02 -0.033 0.973581
## masts2 3.585e-01 1.883e-01 1.904 0.056954 .
## masts3 7.867e-01 1.425e-01 5.522 3.36e-08 ***
## masts4 6.563e-01 2.385e-01 2.751 0.005935 **
## masts5 2.087e-01 1.569e-01 1.330 0.183459
## masts6 1.966e-01 2.048e-01 0.960 0.337170
## masts77 2.931e+00 1.428e+00 2.052 0.040210 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 3101.7 on 4878 degrees of freedom
## Residual deviance: 2858.6 on 4859 degrees of freedom
## (493 observations deleted due to missingness)
## AIC: 2898.6
##
## Number of Fisher Scoring iterations: 11logistic.display(Model2_2)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## sHCOT (cont. var.) 1.004 (1.0028,1.0052) 1.0031 (1.0018,1.0043)
##
## age (cont. var.) 1.0074 (1.002,1.0129) 1.0033 (0.9962,1.0104)
##
## sex: 2 vs 1 2.07 (1.69,2.54) 2.05 (1.66,2.54)
##
## ethnic: ref.=1
## 2 1.29 (0.88,1.88) 1.37 (0.92,2.03)
## 3 0.96 (0.72,1.29) 1.2 (0.86,1.68)
## 4 1.06 (0.77,1.47) 1.15 (0.8,1.65)
## 6 0.29 (0.17,0.51) 0.51 (0.28,0.9)
## 7 1.57 (0.94,2.62) 2.11 (1.23,3.64)
##
## educ: ref.=1
## 2 0.85 (0.6,1.21) 0.77 (0.53,1.12)
## 3 0.54 (0.38,0.76) 0.48 (0.33,0.7)
## 4 0.49 (0.36,0.68) 0.43 (0.3,0.63)
## 5 0.21 (0.14,0.31) 0.24 (0.15,0.36)
## 9 0 (0,4.35136067215608e+271) 0 (0,5.64549826606785e+271)
##
## masts: ref.=1
## 2 2.3 (1.66,3.19) 1.43 (0.99,2.07)
## 3 2.67 (2.04,3.49) 2.2 (1.66,2.9)
## 4 2.95 (1.88,4.63) 1.93 (1.21,3.08)
## 5 1.26 (0.96,1.66) 1.23 (0.91,1.68)
## 6 1.55 (1.06,2.25) 1.22 (0.81,1.82)
## 77 13.14 (0.82,211) 18.74 (1.14,308.02)
##
## P(Wald's test) P(LR-test)
## sHCOT (cont. var.) < 0.001 < 0.001
##
## age (cont. var.) 0.36 0.361
##
## sex: 2 vs 1 < 0.001 < 0.001
##
## ethnic: ref.=1 < 0.001
## 2 0.121
## 3 0.281
## 4 0.459
## 6 0.021
## 7 0.007
##
## educ: ref.=1 < 0.001
## 2 0.17
## 3 < 0.001
## 4 < 0.001
## 5 < 0.001
## 9 0.974
##
## masts: ref.=1 < 0.001
## 2 0.057
## 3 < 0.001
## 4 0.006
## 5 0.183
## 6 0.337
## 77 0.04
##
## Log-likelihood = -1429.318
## No. of observations = 4879
## AIC value = 2898.6359summary(Model2_3)##
## Call:
## glm(formula = dp10 ~ uCOT + age + sex + ethnic + educ + masts,
## family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.6941 -0.4599 -0.3359 -0.2395 3.0598
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.687e+00 5.137e-01 -5.230 1.69e-07 ***
## uCOT 1.230e-04 3.613e-05 3.404 0.000664 ***
## age 4.480e-04 6.960e-03 0.064 0.948678
## sex2 7.430e-01 2.021e-01 3.676 0.000237 ***
## ethnic2 1.622e-01 4.069e-01 0.399 0.690201
## ethnic3 4.823e-01 3.338e-01 1.445 0.148556
## ethnic4 6.096e-01 3.569e-01 1.708 0.087605 .
## ethnic6 6.524e-02 4.847e-01 0.135 0.892937
## ethnic7 8.495e-01 5.461e-01 1.556 0.119798
## educ2 -3.808e-01 3.628e-01 -1.050 0.293829
## educ3 -7.574e-01 3.535e-01 -2.142 0.032162 *
## educ4 -9.327e-01 3.494e-01 -2.669 0.007605 **
## educ5 -2.082e+00 4.537e-01 -4.588 4.46e-06 ***
## masts2 2.385e-01 3.739e-01 0.638 0.523656
## masts3 8.097e-01 2.616e-01 3.095 0.001967 **
## masts4 8.101e-01 4.422e-01 1.832 0.066960 .
## masts5 -9.335e-03 2.993e-01 -0.031 0.975120
## masts6 2.498e-01 3.635e-01 0.687 0.491879
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 923.16 on 1611 degrees of freedom
## Residual deviance: 836.36 on 1594 degrees of freedom
## (3760 observations deleted due to missingness)
## AIC: 872.36
##
## Number of Fisher Scoring iterations: 6logistic.display(Model2_3)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## uCOT (cont. var.) 1.0001 (1.0001,1.0002) 1.0001 (1.0001,1.0002)
##
## age (cont. var.) 1.0037 (0.9935,1.014) 1.0004 (0.9869,1.0142)
##
## sex: 2 vs 1 1.91 (1.32,2.76) 2.1 (1.41,3.12)
##
## ethnic: ref.=1
## 2 1.14 (0.53,2.44) 1.18 (0.53,2.61)
## 3 1.11 (0.64,1.92) 1.62 (0.84,3.12)
## 4 1.35 (0.74,2.48) 1.84 (0.91,3.7)
## 6 0.51 (0.21,1.25) 1.07 (0.41,2.76)
## 7 1.81 (0.68,4.84) 2.34 (0.8,6.82)
##
## educ: ref.=1
## 2 1.02 (0.54,1.94) 0.68 (0.34,1.39)
## 3 0.72 (0.39,1.32) 0.47 (0.23,0.94)
## 4 0.58 (0.33,1.05) 0.39 (0.2,0.78)
## 5 0.16 (0.07,0.35) 0.12 (0.05,0.3)
##
## masts: ref.=1
## 2 2.05 (1.05,3.99) 1.27 (0.61,2.64)
## 3 3.03 (1.86,4.91) 2.25 (1.35,3.75)
## 4 3.36 (1.49,7.6) 2.25 (0.94,5.35)
## 5 1.11 (0.6609,1.8643) 0.9907 (0.551,1.7813)
## 6 1.68 (0.87,3.25) 1.28 (0.63,2.62)
##
## P(Wald's test) P(LR-test)
## uCOT (cont. var.) < 0.001 < 0.001
##
## age (cont. var.) 0.949 0.949
##
## sex: 2 vs 1 < 0.001 < 0.001
##
## ethnic: ref.=1 0.412
## 2 0.69
## 3 0.149
## 4 0.088
## 6 0.893
## 7 0.12
##
## educ: ref.=1 < 0.001
## 2 0.294
## 3 0.032
## 4 0.008
## 5 < 0.001
##
## masts: ref.=1 0.036
## 2 0.524
## 3 0.002
## 4 0.067
## 5 0.975
## 6 0.492
##
## Log-likelihood = -418.1821
## No. of observations = 1612
## AIC value = 872.3642summary(Model2_4)##
## Call:
## glm(formula = dp10 ~ uHCOT + age + sex + ethnic + educ + masts,
## family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.6126 -0.4636 -0.3403 -0.2409 3.0621
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.584e+00 5.093e-01 -5.074 3.90e-07 ***
## uHCOT 4.243e-05 1.532e-05 2.770 0.005614 **
## age -1.258e-03 6.866e-03 -0.183 0.854636
## sex2 7.151e-01 2.009e-01 3.559 0.000372 ***
## ethnic2 1.877e-01 4.060e-01 0.462 0.643945
## ethnic3 5.670e-01 3.300e-01 1.718 0.085794 .
## ethnic4 6.340e-01 3.564e-01 1.779 0.075292 .
## ethnic6 1.192e-01 4.838e-01 0.246 0.805412
## ethnic7 9.166e-01 5.441e-01 1.685 0.092046 .
## educ2 -3.651e-01 3.619e-01 -1.009 0.313075
## educ3 -7.537e-01 3.530e-01 -2.135 0.032720 *
## educ4 -9.501e-01 3.491e-01 -2.722 0.006491 **
## educ5 -2.126e+00 4.535e-01 -4.688 2.75e-06 ***
## masts2 2.442e-01 3.738e-01 0.653 0.513478
## masts3 8.099e-01 2.610e-01 3.103 0.001917 **
## masts4 8.409e-01 4.432e-01 1.897 0.057771 .
## masts5 -3.404e-02 2.987e-01 -0.114 0.909263
## masts6 2.483e-01 3.628e-01 0.684 0.493720
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 923.16 on 1611 degrees of freedom
## Residual deviance: 840.18 on 1594 degrees of freedom
## (3760 observations deleted due to missingness)
## AIC: 876.18
##
## Number of Fisher Scoring iterations: 6logistic.display(Model2_4)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## uHCOT (cont. var.) 1.0001 (1,1.0001) 1 (1,1.0001)
##
## age (cont. var.) 1.0037 (0.9935,1.014) 0.9987 (0.9854,1.0123)
##
## sex: 2 vs 1 1.91 (1.32,2.76) 2.04 (1.38,3.03)
##
## ethnic: ref.=1
## 2 1.14 (0.53,2.44) 1.21 (0.54,2.67)
## 3 1.11 (0.64,1.92) 1.76 (0.92,3.37)
## 4 1.35 (0.74,2.48) 1.89 (0.94,3.79)
## 6 0.51 (0.21,1.25) 1.13 (0.44,2.91)
## 7 1.81 (0.68,4.84) 2.5 (0.86,7.26)
##
## educ: ref.=1
## 2 1.02 (0.54,1.94) 0.69 (0.34,1.41)
## 3 0.72 (0.39,1.32) 0.47 (0.24,0.94)
## 4 0.58 (0.33,1.05) 0.39 (0.2,0.77)
## 5 0.16 (0.07,0.35) 0.12 (0.05,0.29)
##
## masts: ref.=1
## 2 2.05 (1.05,3.99) 1.28 (0.61,2.66)
## 3 3.03 (1.86,4.91) 2.25 (1.35,3.75)
## 4 3.36 (1.49,7.6) 2.32 (0.97,5.53)
## 5 1.11 (0.66,1.86) 0.97 (0.54,1.74)
## 6 1.68 (0.87,3.25) 1.28 (0.63,2.61)
##
## P(Wald's test) P(LR-test)
## uHCOT (cont. var.) 0.006 0.007
##
## age (cont. var.) 0.855 0.855
##
## sex: 2 vs 1 < 0.001 < 0.001
##
## ethnic: ref.=1 0.33
## 2 0.644
## 3 0.086
## 4 0.075
## 6 0.805
## 7 0.092
##
## educ: ref.=1 < 0.001
## 2 0.313
## 3 0.033
## 4 0.006
## 5 < 0.001
##
## masts: ref.=1 0.03
## 2 0.513
## 3 0.002
## 4 0.058
## 5 0.909
## 6 0.494
##
## Log-likelihood = -420.0923
## No. of observations = 1612
## AIC value = 876.1845# Plot forest
forest_Model2_1 = ggcoef(Model2_1, exponentiate=T, exclude_intercept=T, vline_color = "red",
errorbar_color = "blue", errorbar_height = 0.10)## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurredforest_Model2_1## Warning: Removed 1 rows containing missing values (`geom_errorbarh()`).
forest_Model2_2 = ggcoef(Model2_2, exponentiate=T, exclude_intercept=T, vline_color = "red",
errorbar_color = "blue", errorbar_height = 0.10)## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurredforest_Model2_2## Warning: Removed 1 rows containing missing values (`geom_errorbarh()`).
forest_Model2_3 = ggcoef(Model2_3, exponentiate=T, exclude_intercept=T, vline_color = "red",
errorbar_color = "blue", errorbar_height = 0.10)
forest_Model2_3
forest_Model2_4 = ggcoef(Model2_4, exponentiate=T, exclude_intercept=T, vline_color = "red",
errorbar_color = "blue", errorbar_height = 0.10)
forest_Model2_4
### Adjust for: age, sex, ethnic group, educ, marrital status, finc, PIR
Model2_1_1 = glm(dp10 ~ sCOT + age + sex + ethnic + educ + masts + finc + PIR, family = binomial, data = tab1)
# Summary the results of logistic regression
summary(Model2_1_1)##
## Call:
## glm(formula = dp10 ~ sCOT + age + sex + ethnic + educ + masts +
## finc + PIR, family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.3297 -0.5007 -0.3551 -0.2398 3.1464
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.800e+00 3.305e-01 -5.447 5.13e-08 ***
## sCOT 1.376e-03 3.284e-04 4.189 2.80e-05 ***
## age 2.156e-03 3.853e-03 0.559 0.57584
## sex2 7.281e-01 1.134e-01 6.418 1.38e-10 ***
## ethnic2 2.444e-01 2.197e-01 1.113 0.26584
## ethnic3 1.651e-01 1.813e-01 0.911 0.36234
## ethnic4 6.857e-02 1.974e-01 0.347 0.72839
## ethnic6 -7.629e-01 3.257e-01 -2.342 0.01916 *
## ethnic7 7.227e-01 2.907e-01 2.487 0.01290 *
## educ2 -2.601e-01 2.067e-01 -1.259 0.20814
## educ3 -5.891e-01 2.058e-01 -2.862 0.00421 **
## educ4 -5.949e-01 2.034e-01 -2.924 0.00346 **
## educ5 -1.026e+00 2.490e-01 -4.119 3.80e-05 ***
## educ9 -1.086e+01 3.247e+02 -0.033 0.97332
## masts2 6.916e-02 2.028e-01 0.341 0.73307
## masts3 4.869e-01 1.567e-01 3.107 0.00189 **
## masts4 2.921e-01 2.582e-01 1.131 0.25794
## masts5 -8.680e-02 1.719e-01 -0.505 0.61352
## masts6 -2.692e-02 2.175e-01 -0.124 0.90148
## masts77 3.124e+00 1.425e+00 2.193 0.02834 *
## finc -1.094e-01 3.548e-02 -3.084 0.00204 **
## PIR 3.413e-02 9.335e-02 0.366 0.71462
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 2856.2 on 4528 degrees of freedom
## Residual deviance: 2590.1 on 4507 degrees of freedom
## (843 observations deleted due to missingness)
## AIC: 2634.1
##
## Number of Fisher Scoring iterations: 11logistic.display(Model2_1_1)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## sCOT (cont. var.) 1.0019 (1.0014,1.0025) 1.0014 (1.0007,1.002)
##
## age (cont. var.) 1.0067 (1.0011,1.0124) 1.0022 (0.9946,1.0098)
##
## sex: 2 vs 1 2.04 (1.65,2.51) 2.07 (1.66,2.59)
##
## ethnic: ref.=1
## 2 1.23 (0.81,1.85) 1.28 (0.83,1.96)
## 3 0.98 (0.72,1.34) 1.18 (0.83,1.68)
## 4 1.07 (0.76,1.52) 1.07 (0.73,1.58)
## 6 0.26 (0.14,0.48) 0.47 (0.25,0.88)
## 7 1.67 (0.98,2.84) 2.06 (1.17,3.64)
##
## educ: ref.=1
## 2 0.86 (0.59,1.26) 0.77 (0.51,1.16)
## 3 0.56 (0.39,0.81) 0.55 (0.37,0.83)
## 4 0.52 (0.37,0.73) 0.55 (0.37,0.82)
## 5 0.21 (0.14,0.32) 0.36 (0.22,0.58)
## 9 0 (0,4.48617189068007e+271) 0 (0,5.08287444546071e+271)
##
## masts: ref.=1
## 2 2.21 (1.56,3.13) 1.07 (0.72,1.59)
## 3 2.72 (2.06,3.58) 1.63 (1.2,2.21)
## 4 2.83 (1.76,4.56) 1.34 (0.81,2.22)
## 5 1.27 (0.95,1.69) 0.92 (0.65,1.28)
## 6 1.63 (1.1,2.4) 0.97 (0.64,1.49)
## 77 13.41 (0.83,215.34) 22.74 (1.39,371.28)
##
## finc (cont. var.) 0.86 (0.84,0.88) 0.9 (0.84,0.96)
##
## PIR (cont. var.) 0.69 (0.65,0.75) 1.03 (0.86,1.24)
##
## P(Wald's test) P(LR-test)
## sCOT (cont. var.) < 0.001 < 0.001
##
## age (cont. var.) 0.576 0.576
##
## sex: 2 vs 1 < 0.001 < 0.001
##
## ethnic: ref.=1 0.002
## 2 0.266
## 3 0.362
## 4 0.728
## 6 0.019
## 7 0.013
##
## educ: ref.=1 < 0.001
## 2 0.208
## 3 0.004
## 4 0.003
## 5 < 0.001
## 9 0.973
##
## masts: ref.=1 0.01
## 2 0.733
## 3 0.002
## 4 0.258
## 5 0.614
## 6 0.901
## 77 0.028
##
## finc (cont. var.) 0.002 0.002
##
## PIR (cont. var.) 0.715 0.715
##
## Log-likelihood = -1295.029
## No. of observations = 4529
## AIC value = 2634.05798.4 Model 3
### Adjust for: age, sex, ethnic group, educ, marrital status + finc + PIR + smoking + alcohol + physical activity
Model3_1 = glm(dp10 ~ sCOT + age + sex + ethnic + educ + masts + finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650, family = binomial, data = tab1)
Model3_2 = glm(dp10 ~ sHCOT + age + sex + ethnic + educ + masts + finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650, family = binomial, data = tab1)
Model3_3 = glm(dp10 ~ uCOT + age + sex + ethnic + educ + masts + finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650, family = binomial, data = tab1)
Model3_4 = glm(dp10 ~ uHCOT + age + sex + ethnic + educ + masts + finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650, family = binomial, data = tab1)
# Summary the results of logistic regression
summary(Model3_1)##
## Call:
## glm(formula = dp10 ~ sCOT + age + sex + ethnic + educ + masts +
## finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650, family = binomial,
## data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.1918 -0.4903 -0.3471 -0.2203 3.2194
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.233e+00 3.762e-01 -5.936 2.93e-09 ***
## sCOT 9.100e-04 3.614e-04 2.518 0.011807 *
## age -1.115e-03 4.012e-03 -0.278 0.781080
## sex2 7.814e-01 1.180e-01 6.621 3.56e-11 ***
## ethnic2 2.336e-01 2.211e-01 1.056 0.290847
## ethnic3 9.024e-02 1.827e-01 0.494 0.621453
## ethnic4 6.395e-02 1.986e-01 0.322 0.747417
## ethnic6 -7.030e-01 3.271e-01 -2.149 0.031636 *
## ethnic7 6.610e-01 2.956e-01 2.236 0.025360 *
## educ2 -2.809e-01 2.087e-01 -1.346 0.178338
## educ3 -5.683e-01 2.082e-01 -2.730 0.006341 **
## educ4 -5.360e-01 2.059e-01 -2.603 0.009243 **
## educ5 -8.629e-01 2.519e-01 -3.426 0.000613 ***
## educ9 -1.197e+01 5.354e+02 -0.022 0.982166
## masts2 1.016e-01 2.042e-01 0.497 0.618939
## masts3 4.718e-01 1.590e-01 2.967 0.003006 **
## masts4 2.602e-01 2.605e-01 0.999 0.317830
## masts5 -9.140e-02 1.737e-01 -0.526 0.598833
## masts6 -9.277e-02 2.199e-01 -0.422 0.673170
## masts77 3.176e+00 1.491e+00 2.130 0.033164 *
## finc -1.063e-01 3.589e-02 -2.962 0.003059 **
## PIR 4.615e-02 9.437e-02 0.489 0.624787
## SMQ0202 -2.672e-01 1.259e-01 -2.123 0.033796 *
## SMQ0209 -1.167e+01 3.767e+02 -0.031 0.975285
## ALQ1012 -2.165e-01 1.326e-01 -1.632 0.102615
## ALQ1019 9.603e-01 8.475e-01 1.133 0.257213
## PAQ6652 5.203e-01 1.210e-01 4.301 1.70e-05 ***
## PAQ6502 4.705e-01 1.757e-01 2.678 0.007397 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 2856.0 on 4527 degrees of freedom
## Residual deviance: 2544.4 on 4500 degrees of freedom
## (844 observations deleted due to missingness)
## AIC: 2600.4
##
## Number of Fisher Scoring iterations: 12logistic.display(Model3_1)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## sCOT (cont. var.) 1.0019 (1.0014,1.0025) 1.0009 (1.0002,1.0016)
##
## age (cont. var.) 1.0068 (1.0011,1.0125) 0.9989 (0.9911,1.0068)
##
## sex: 2 vs 1 2.04 (1.65,2.51) 2.18 (1.73,2.75)
##
## ethnic: ref.=1
## 2 1.23 (0.81,1.85) 1.26 (0.82,1.95)
## 3 0.99 (0.72,1.34) 1.09 (0.76,1.57)
## 4 1.07 (0.76,1.52) 1.07 (0.72,1.57)
## 6 0.26 (0.14,0.48) 0.5 (0.26,0.94)
## 7 1.67 (0.98,2.84) 1.94 (1.08,3.46)
##
## educ: ref.=1
## 2 0.86 (0.59,1.26) 0.76 (0.5,1.14)
## 3 0.56 (0.39,0.81) 0.57 (0.38,0.85)
## 4 0.52 (0.37,0.73) 0.59 (0.39,0.88)
## 5 0.21 (0.14,0.32) 0.42 (0.26,0.69)
## 9 0 (0,4.48617189182557e+271) 0 (0,Inf)
##
## masts: ref.=1
## 2 2.21 (1.56,3.13) 1.11 (0.74,1.65)
## 3 2.72 (2.06,3.58) 1.6 (1.17,2.19)
## 4 2.83 (1.76,4.56) 1.3 (0.78,2.16)
## 5 1.27 (0.95,1.69) 0.91 (0.65,1.28)
## 6 1.63 (1.1,2.4) 0.91 (0.59,1.4)
## 77 13.4 (0.83,215.25) 23.94 (1.29,444.72)
##
## finc (cont. var.) 0.86 (0.84,0.88) 0.9 (0.84,0.96)
##
## PIR (cont. var.) 0.69 (0.65,0.75) 1.05 (0.87,1.26)
##
## SMQ020: ref.=1
## 2 0.54 (0.44,0.66) 0.77 (0.6,0.98)
## 9 0 (0,6.98824325679386e+190) 0 (0,Inf)
##
## ALQ101: ref.=1
## 2 0.96 (0.76,1.2) 0.81 (0.62,1.04)
## 9 3.12 (0.63,15.54) 2.61 (0.5,13.75)
##
## PAQ665: 2 vs 1 2.24 (1.79,2.8) 1.68 (1.33,2.13)
##
## PAQ650: 2 vs 1 2.75 (2.01,3.78) 1.6 (1.13,2.26)
##
## P(Wald's test) P(LR-test)
## sCOT (cont. var.) 0.012 0.014
##
## age (cont. var.) 0.781 0.781
##
## sex: 2 vs 1 < 0.001 < 0.001
##
## ethnic: ref.=1 0.01
## 2 0.291
## 3 0.621
## 4 0.747
## 6 0.032
## 7 0.025
##
## educ: ref.=1 0.014
## 2 0.178
## 3 0.006
## 4 0.009
## 5 < 0.001
## 9 0.982
##
## masts: ref.=1 0.015
## 2 0.619
## 3 0.003
## 4 0.318
## 5 0.599
## 6 0.673
## 77 0.033
##
## finc (cont. var.) 0.003 0.003
##
## PIR (cont. var.) 0.625 0.625
##
## SMQ020: ref.=1 0.081
## 2 0.034
## 9 0.975
##
## ALQ101: ref.=1 0.138
## 2 0.103
## 9 0.257
##
## PAQ665: 2 vs 1 < 0.001 < 0.001
##
## PAQ650: 2 vs 1 0.007 0.005
##
## Log-likelihood = -1272.2219
## No. of observations = 4528
## AIC value = 2600.4438summary(Model3_2)##
## Call:
## glm(formula = dp10 ~ sHCOT + age + sex + ethnic + educ + masts +
## finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650, family = binomial,
## data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.2018 -0.4888 -0.3477 -0.2214 3.2088
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.182e+00 3.735e-01 -5.841 5.19e-09 ***
## sHCOT 1.623e-03 6.863e-04 2.364 0.018059 *
## age -1.724e-03 3.993e-03 -0.432 0.665909
## sex2 7.668e-01 1.176e-01 6.519 7.08e-11 ***
## ethnic2 2.379e-01 2.209e-01 1.077 0.281700
## ethnic3 9.809e-02 1.825e-01 0.538 0.590891
## ethnic4 9.160e-02 1.969e-01 0.465 0.641697
## ethnic6 -6.873e-01 3.268e-01 -2.103 0.035457 *
## ethnic7 6.776e-01 2.951e-01 2.296 0.021677 *
## educ2 -2.695e-01 2.086e-01 -1.292 0.196401
## educ3 -5.626e-01 2.082e-01 -2.702 0.006883 **
## educ4 -5.366e-01 2.059e-01 -2.606 0.009166 **
## educ5 -8.669e-01 2.519e-01 -3.442 0.000578 ***
## educ9 -1.198e+01 5.354e+02 -0.022 0.982154
## masts2 9.408e-02 2.045e-01 0.460 0.645461
## masts3 4.687e-01 1.591e-01 2.945 0.003227 **
## masts4 2.706e-01 2.604e-01 1.039 0.298616
## masts5 -8.383e-02 1.737e-01 -0.483 0.629344
## masts6 -8.817e-02 2.199e-01 -0.401 0.688491
## masts77 3.172e+00 1.498e+00 2.117 0.034226 *
## finc -1.059e-01 3.591e-02 -2.949 0.003192 **
## PIR 4.513e-02 9.445e-02 0.478 0.632807
## SMQ0202 -2.946e-01 1.232e-01 -2.392 0.016749 *
## SMQ0209 -1.169e+01 3.770e+02 -0.031 0.975256
## ALQ1012 -2.116e-01 1.326e-01 -1.596 0.110593
## ALQ1019 9.496e-01 8.486e-01 1.119 0.263134
## PAQ6652 5.251e-01 1.209e-01 4.342 1.41e-05 ***
## PAQ6502 4.719e-01 1.756e-01 2.686 0.007222 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 2856.0 on 4527 degrees of freedom
## Residual deviance: 2545.2 on 4500 degrees of freedom
## (844 observations deleted due to missingness)
## AIC: 2601.2
##
## Number of Fisher Scoring iterations: 12logistic.display(Model3_2)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## sHCOT (cont. var.) 1.0039 (1.0027,1.0051) 1.0016 (1.0003,1.003)
##
## age (cont. var.) 1.0068 (1.0011,1.0125) 0.9983 (0.9905,1.0061)
##
## sex: 2 vs 1 2.04 (1.65,2.51) 2.15 (1.71,2.71)
##
## ethnic: ref.=1
## 2 1.23 (0.81,1.85) 1.27 (0.82,1.96)
## 3 0.99 (0.72,1.34) 1.1 (0.77,1.58)
## 4 1.07 (0.76,1.52) 1.1 (0.75,1.61)
## 6 0.26 (0.14,0.48) 0.5 (0.27,0.95)
## 7 1.67 (0.98,2.84) 1.97 (1.1,3.51)
##
## educ: ref.=1
## 2 0.86 (0.59,1.26) 0.76 (0.51,1.15)
## 3 0.56 (0.39,0.81) 0.57 (0.38,0.86)
## 4 0.52 (0.37,0.73) 0.58 (0.39,0.88)
## 5 0.21 (0.14,0.32) 0.42 (0.26,0.69)
## 9 0 (0,4.48617189182557e+271) 0 (0,Inf)
##
## masts: ref.=1
## 2 2.21 (1.56,3.13) 1.1 (0.74,1.64)
## 3 2.72 (2.06,3.58) 1.6 (1.17,2.18)
## 4 2.83 (1.76,4.56) 1.31 (0.79,2.18)
## 5 1.27 (0.95,1.69) 0.92 (0.65,1.29)
## 6 1.63 (1.1,2.4) 0.92 (0.59,1.41)
## 77 13.4 (0.83,215.25) 23.85 (1.27,449.2)
##
## finc (cont. var.) 0.86 (0.84,0.88) 0.9 (0.84,0.97)
##
## PIR (cont. var.) 0.69 (0.65,0.75) 1.05 (0.87,1.26)
##
## SMQ020: ref.=1
## 2 0.54 (0.44,0.66) 0.74 (0.59,0.95)
## 9 0 (0,6.98824325679386e+190) 0 (0,Inf)
##
## ALQ101: ref.=1
## 2 0.96 (0.76,1.2) 0.81 (0.62,1.05)
## 9 3.12 (0.63,15.54) 2.58 (0.49,13.64)
##
## PAQ665: 2 vs 1 2.24 (1.79,2.8) 1.69 (1.33,2.14)
##
## PAQ650: 2 vs 1 2.75 (2.01,3.78) 1.6 (1.14,2.26)
##
## P(Wald's test) P(LR-test)
## sHCOT (cont. var.) 0.018 0.022
##
## age (cont. var.) 0.666 0.666
##
## sex: 2 vs 1 < 0.001 < 0.001
##
## ethnic: ref.=1 0.011
## 2 0.282
## 3 0.591
## 4 0.642
## 6 0.035
## 7 0.022
##
## educ: ref.=1 0.012
## 2 0.196
## 3 0.007
## 4 0.009
## 5 < 0.001
## 9 0.982
##
## masts: ref.=1 0.016
## 2 0.645
## 3 0.003
## 4 0.299
## 5 0.629
## 6 0.688
## 77 0.034
##
## finc (cont. var.) 0.003 0.003
##
## PIR (cont. var.) 0.633 0.633
##
## SMQ020: ref.=1 0.044
## 2 0.017
## 9 0.975
##
## ALQ101: ref.=1 0.149
## 2 0.111
## 9 0.263
##
## PAQ665: 2 vs 1 < 0.001 < 0.001
##
## PAQ650: 2 vs 1 0.007 0.005
##
## Log-likelihood = -1272.618
## No. of observations = 4528
## AIC value = 2601.236summary(Model3_3)##
## Call:
## glm(formula = dp10 ~ uCOT + age + sex + ethnic + educ + masts +
## finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650, family = binomial,
## data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.3379 -0.4374 -0.3016 -0.1764 3.0384
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.790e+00 7.284e-01 -2.457 0.01401 *
## uCOT 1.092e-04 3.758e-05 2.907 0.00365 **
## age -4.794e-03 7.712e-03 -0.622 0.53417
## sex2 9.486e-01 2.250e-01 4.215 2.5e-05 ***
## ethnic2 -9.306e-02 4.646e-01 -0.200 0.84123
## ethnic3 1.960e-01 3.640e-01 0.538 0.59031
## ethnic4 3.789e-01 3.875e-01 0.978 0.32809
## ethnic6 -1.250e-01 6.091e-01 -0.205 0.83739
## ethnic7 5.081e-01 5.764e-01 0.882 0.37800
## educ2 -3.633e-01 4.225e-01 -0.860 0.38996
## educ3 -4.660e-01 4.078e-01 -1.143 0.25316
## educ4 -4.845e-01 4.078e-01 -1.188 0.23486
## educ5 -1.324e+00 5.179e-01 -2.557 0.01057 *
## masts2 5.711e-02 4.059e-01 0.141 0.88811
## masts3 4.540e-01 2.926e-01 1.552 0.12069
## masts4 1.095e-02 5.510e-01 0.020 0.98415
## masts5 -3.056e-01 3.299e-01 -0.926 0.35438
## masts6 -1.623e-01 3.953e-01 -0.411 0.68135
## finc -1.750e-01 6.979e-02 -2.508 0.01215 *
## PIR 1.427e-01 1.734e-01 0.823 0.41058
## SMQ0202 -2.785e-01 2.304e-01 -1.209 0.22679
## ALQ1012 -5.690e-01 2.673e-01 -2.128 0.03331 *
## ALQ1019 2.613e+00 1.574e+00 1.660 0.09688 .
## PAQ6652 4.716e-01 2.272e-01 2.076 0.03788 *
## PAQ6502 1.874e-01 3.214e-01 0.583 0.55989
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 841.69 on 1502 degrees of freedom
## Residual deviance: 726.01 on 1478 degrees of freedom
## (3869 observations deleted due to missingness)
## AIC: 776.01
##
## Number of Fisher Scoring iterations: 6logistic.display(Model3_3)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## uCOT (cont. var.) 1.0001 (1.0001,1.0002) 1.0001 (1,1.0002)
##
## age (cont. var.) 1.0027 (0.992,1.0135) 0.9952 (0.9803,1.0104)
##
## sex: 2 vs 1 1.97 (1.34,2.91) 2.58 (1.66,4.01)
##
## ethnic: ref.=1
## 2 1.01 (0.43,2.39) 0.91 (0.37,2.26)
## 3 1.2 (0.67,2.16) 1.22 (0.6,2.48)
## 4 1.37 (0.71,2.64) 1.46 (0.68,3.12)
## 6 0.33 (0.11,1.02) 0.88 (0.27,2.91)
## 7 2.01 (0.73,5.51) 1.66 (0.54,5.14)
##
## educ: ref.=1
## 2 1.08 (0.52,2.24) 0.7 (0.3,1.59)
## 3 0.89 (0.45,1.76) 0.63 (0.28,1.4)
## 4 0.73 (0.38,1.42) 0.62 (0.28,1.37)
## 5 0.2 (0.08,0.46) 0.27 (0.1,0.73)
##
## masts: ref.=1
## 2 2.16 (1.08,4.35) 1.06 (0.48,2.35)
## 3 3.3 (1.99,5.46) 1.57 (0.89,2.79)
## 4 2.46 (0.92,6.6) 1.01 (0.34,2.98)
## 5 1.28 (0.75,2.17) 0.74 (0.39,1.41)
## 6 1.69 (0.84,3.37) 0.85 (0.39,1.84)
##
## finc (cont. var.) 0.84 (0.8,0.89) 0.84 (0.73,0.96)
##
## PIR (cont. var.) 0.67 (0.58,0.76) 1.15 (0.82,1.62)
##
## SMQ020: 2 vs 1 0.46 (0.32,0.68) 0.76 (0.48,1.19)
##
## ALQ101: ref.=1
## 2 0.7 (0.44,1.11) 0.57 (0.34,0.96)
## 9 10.58 (0.66,170.54) 13.65 (0.62,298.53)
##
## PAQ665: 2 vs 1 1.99 (1.32,2.99) 1.6 (1.03,2.5)
##
## PAQ650: 2 vs 1 2.35 (1.33,4.16) 1.21 (0.64,2.26)
##
## P(Wald's test) P(LR-test)
## uCOT (cont. var.) 0.004 0.007
##
## age (cont. var.) 0.534 0.534
##
## sex: 2 vs 1 < 0.001 < 0.001
##
## ethnic: ref.=1 0.791
## 2 0.841
## 3 0.59
## 4 0.328
## 6 0.837
## 7 0.378
##
## educ: ref.=1 0.085
## 2 0.39
## 3 0.253
## 4 0.235
## 5 0.011
##
## masts: ref.=1 0.368
## 2 0.888
## 3 0.121
## 4 0.984
## 5 0.354
## 6 0.681
##
## finc (cont. var.) 0.012 0.01
##
## PIR (cont. var.) 0.411 0.412
##
## SMQ020: 2 vs 1 0.227 0.225
##
## ALQ101: ref.=1 0.024
## 2 0.033
## 9 0.097
##
## PAQ665: 2 vs 1 0.038 0.034
##
## PAQ650: 2 vs 1 0.56 0.555
##
## Log-likelihood = -363.006
## No. of observations = 1503
## AIC value = 776.012summary(Model3_4)##
## Call:
## glm(formula = dp10 ~ uHCOT + age + sex + ethnic + educ + masts +
## finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650, family = binomial,
## data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.3394 -0.4427 -0.3030 -0.1762 3.0726
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.713e+00 7.265e-01 -2.358 0.01838 *
## uHCOT 3.749e-05 1.593e-05 2.354 0.01858 *
## age -6.734e-03 7.597e-03 -0.886 0.37538
## sex2 9.217e-01 2.236e-01 4.123 3.74e-05 ***
## ethnic2 -7.080e-02 4.637e-01 -0.153 0.87865
## ethnic3 2.750e-01 3.598e-01 0.764 0.44460
## ethnic4 4.026e-01 3.871e-01 1.040 0.29827
## ethnic6 -7.212e-02 6.082e-01 -0.119 0.90561
## ethnic7 5.624e-01 5.754e-01 0.977 0.32841
## educ2 -3.392e-01 4.211e-01 -0.805 0.42053
## educ3 -4.550e-01 4.064e-01 -1.120 0.26278
## educ4 -4.904e-01 4.067e-01 -1.206 0.22788
## educ5 -1.340e+00 5.179e-01 -2.587 0.00969 **
## masts2 6.109e-02 4.061e-01 0.150 0.88041
## masts3 4.515e-01 2.925e-01 1.544 0.12263
## masts4 3.781e-02 5.509e-01 0.069 0.94529
## masts5 -3.373e-01 3.297e-01 -1.023 0.30627
## masts6 -1.728e-01 3.944e-01 -0.438 0.66125
## finc -1.707e-01 6.979e-02 -2.446 0.01445 *
## PIR 1.295e-01 1.732e-01 0.748 0.45474
## SMQ0202 -3.106e-01 2.300e-01 -1.350 0.17688
## ALQ1012 -5.580e-01 2.670e-01 -2.089 0.03666 *
## ALQ1019 2.762e+00 1.531e+00 1.804 0.07118 .
## PAQ6652 4.853e-01 2.270e-01 2.138 0.03252 *
## PAQ6502 2.247e-01 3.207e-01 0.701 0.48358
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 841.69 on 1502 degrees of freedom
## Residual deviance: 728.67 on 1478 degrees of freedom
## (3869 observations deleted due to missingness)
## AIC: 778.67
##
## Number of Fisher Scoring iterations: 6logistic.display(Model3_4)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## uHCOT (cont. var.) 1.0001 (1,1.0001) 1 (1,1.0001)
##
## age (cont. var.) 1.0027 (0.992,1.0135) 0.9933 (0.9786,1.0082)
##
## sex: 2 vs 1 1.97 (1.34,2.91) 2.51 (1.62,3.9)
##
## ethnic: ref.=1
## 2 1.01 (0.43,2.39) 0.93 (0.38,2.31)
## 3 1.2 (0.67,2.16) 1.32 (0.65,2.66)
## 4 1.37 (0.71,2.64) 1.5 (0.7,3.19)
## 6 0.33 (0.11,1.02) 0.93 (0.28,3.06)
## 7 2.01 (0.73,5.51) 1.75 (0.57,5.42)
##
## educ: ref.=1
## 2 1.08 (0.52,2.24) 0.71 (0.31,1.63)
## 3 0.89 (0.45,1.76) 0.63 (0.29,1.41)
## 4 0.73 (0.38,1.42) 0.61 (0.28,1.36)
## 5 0.2 (0.08,0.46) 0.26 (0.09,0.72)
##
## masts: ref.=1
## 2 2.16 (1.08,4.35) 1.06 (0.48,2.36)
## 3 3.3 (1.99,5.46) 1.57 (0.89,2.79)
## 4 2.46 (0.92,6.6) 1.04 (0.35,3.06)
## 5 1.28 (0.75,2.17) 0.71 (0.37,1.36)
## 6 1.69 (0.84,3.37) 0.84 (0.39,1.82)
##
## finc (cont. var.) 0.84 (0.8,0.89) 0.84 (0.74,0.97)
##
## PIR (cont. var.) 0.67 (0.58,0.76) 1.14 (0.81,1.6)
##
## SMQ020: 2 vs 1 0.46 (0.32,0.68) 0.73 (0.47,1.15)
##
## ALQ101: ref.=1
## 2 0.7 (0.44,1.11) 0.57 (0.34,0.97)
## 9 10.58 (0.66,170.54) 15.83 (0.79,318.09)
##
## PAQ665: 2 vs 1 1.99 (1.32,2.99) 1.62 (1.04,2.53)
##
## PAQ650: 2 vs 1 2.35 (1.33,4.16) 1.25 (0.67,2.35)
##
## P(Wald's test) P(LR-test)
## uHCOT (cont. var.) 0.019 0.031
##
## age (cont. var.) 0.375 0.374
##
## sex: 2 vs 1 < 0.001 < 0.001
##
## ethnic: ref.=1 0.769
## 2 0.879
## 3 0.445
## 4 0.298
## 6 0.906
## 7 0.328
##
## educ: ref.=1 0.075
## 2 0.421
## 3 0.263
## 4 0.228
## 5 0.01
##
## masts: ref.=1 0.332
## 2 0.88
## 3 0.123
## 4 0.945
## 5 0.306
## 6 0.661
##
## finc (cont. var.) 0.014 0.012
##
## PIR (cont. var.) 0.455 0.456
##
## SMQ020: 2 vs 1 0.177 0.175
##
## ALQ101: ref.=1 0.023
## 2 0.037
## 9 0.071
##
## PAQ665: 2 vs 1 0.033 0.029
##
## PAQ650: 2 vs 1 0.484 0.476
##
## Log-likelihood = -364.3357
## No. of observations = 1503
## AIC value = 778.6713# Plot forest
forest_Model3_1 = ggcoef(Model3_1, exponentiate=T, exclude_intercept=T, vline_color = "red",
errorbar_color = "blue", errorbar_height = 0.10)## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurredforest_Model3_1## Warning: Removed 2 rows containing missing values (`geom_errorbarh()`).
forest_Model3_2 = ggcoef(Model3_2, exponentiate=T, exclude_intercept=T, vline_color = "red",
errorbar_color = "blue", errorbar_height = 0.10)## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurredforest_Model3_2## Warning: Removed 2 rows containing missing values (`geom_errorbarh()`).
forest_Model3_3 = ggcoef(Model3_3, exponentiate=T, exclude_intercept=T, vline_color = "red",
errorbar_color = "blue", errorbar_height = 0.10)
forest_Model3_3
forest_Model3_4 = ggcoef(Model3_4, exponentiate=T, exclude_intercept=T, vline_color = "red",
errorbar_color = "blue", errorbar_height = 0.10)
forest_Model3_4
8.5 Model 4
### Adjust for: age, sex, ethnic group, educ, marrital status + finc + PIR + smoking + alcohol + physical activity + BMI + SBP + DBP + AST + ALT + BUN + Crea + TC + TG + HDL + LDL + ApoB
Model4_1 = glm(dp10 ~ sCOT + age + sex + ethnic + educ + masts + finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 + BMI + SBP + DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB, family = binomial, data = tab1)
Model4_2 = glm(dp10 ~ sHCOT + age + sex + ethnic + educ + masts + finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 +BMI + SBP + DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB, family = binomial, data = tab1)
Model4_3 = glm(dp10 ~ uCOT + age + sex + ethnic + educ + masts + finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 + BMI + SBP + DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB, family = binomial, data = tab1)
Model4_4 = glm(dp10 ~ uHCOT + age + sex + ethnic + educ + masts + finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 + BMI + SBP + DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB, family = binomial, data = tab1)
# Summary the results of logistic regression
summary(Model4_1)##
## Call:
## glm(formula = dp10 ~ sCOT + age + sex + ethnic + educ + masts +
## finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 + BMI + SBP +
## DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB,
## family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.5728 -0.4601 -0.3057 -0.1827 3.3954
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -5.487e+00 1.043e+00 -5.261 1.43e-07 ***
## sCOT 1.045e-03 5.634e-04 1.855 0.063664 .
## age 5.114e-03 7.485e-03 0.683 0.494423
## sex2 5.876e-01 2.094e-01 2.806 0.005022 **
## ethnic2 2.680e-01 3.527e-01 0.760 0.447227
## ethnic3 1.287e-01 2.892e-01 0.445 0.656257
## ethnic4 8.379e-02 3.364e-01 0.249 0.803298
## ethnic6 -5.808e-01 5.335e-01 -1.089 0.276333
## ethnic7 8.288e-02 6.315e-01 0.131 0.895596
## educ2 -1.729e-01 3.483e-01 -0.496 0.619681
## educ3 -4.452e-01 3.498e-01 -1.273 0.203156
## educ4 -1.878e-01 3.381e-01 -0.555 0.578622
## educ5 -4.198e-01 4.090e-01 -1.026 0.304714
## educ9 -1.383e+01 8.827e+02 -0.016 0.987501
## masts2 2.090e-01 3.212e-01 0.651 0.515368
## masts3 4.968e-01 2.485e-01 1.999 0.045572 *
## masts4 1.148e-01 4.369e-01 0.263 0.792768
## masts5 -2.976e-01 2.969e-01 -1.002 0.316107
## masts6 -2.532e-01 3.630e-01 -0.698 0.485352
## finc -9.181e-02 5.722e-02 -1.604 0.108615
## PIR -8.402e-02 1.502e-01 -0.559 0.576013
## SMQ0202 -1.426e-01 1.950e-01 -0.731 0.464490
## SMQ0209 -1.259e+01 8.827e+02 -0.014 0.988623
## ALQ1012 -1.826e-01 2.120e-01 -0.861 0.389174
## ALQ1019 -1.202e+01 4.362e+02 -0.028 0.978023
## PAQ6652 5.995e-01 1.967e-01 3.047 0.002312 **
## PAQ6502 4.203e-01 2.953e-01 1.423 0.154682
## BMI 4.337e-02 1.138e-02 3.811 0.000139 ***
## SBP -3.509e-03 5.589e-03 -0.628 0.530098
## DBP 1.099e-03 7.360e-03 0.149 0.881337
## AST 1.687e-03 3.576e-03 0.472 0.637054
## ALT 3.549e-03 6.158e-03 0.576 0.564401
## BUN 1.352e-02 1.648e-02 0.820 0.412054
## crea 1.257e-01 1.833e-01 0.686 0.492999
## TC -1.626e-01 2.988e-01 -0.544 0.586309
## TG 3.698e-02 5.973e-02 0.619 0.535894
## HDL 1.731e-01 2.987e-01 0.579 0.562338
## LDL 1.596e-01 2.986e-01 0.534 0.593024
## apoB 7.649e-03 1.017e-02 0.752 0.452185
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1170.32 on 1935 degrees of freedom
## Residual deviance: 997.34 on 1897 degrees of freedom
## (3436 observations deleted due to missingness)
## AIC: 1075.3
##
## Number of Fisher Scoring iterations: 13logistic.display(Model4_1)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## sCOT (cont. var.) 1.0015 (1.0006,1.0024) 1.001 (0.9999,1.0022)
##
## age (cont. var.) 1.0151 (1.006,1.0242) 1.0051 (0.9905,1.02)
##
## sex: 2 vs 1 1.75 (1.26,2.41) 1.8 (1.19,2.71)
##
## ethnic: ref.=1
## 2 1.16 (0.62,2.2) 1.31 (0.65,2.61)
## 3 1 (0.62,1.61) 1.14 (0.65,2)
## 4 0.89 (0.51,1.55) 1.09 (0.56,2.1)
## 6 0.22 (0.08,0.59) 0.56 (0.2,1.59)
## 7 0.89 (0.29,2.7) 1.09 (0.32,3.75)
##
## educ: ref.=1
## 2 0.66 (0.36,1.19) 0.84 (0.43,1.67)
## 3 0.45 (0.25,0.81) 0.64 (0.32,1.27)
## 4 0.52 (0.31,0.9) 0.83 (0.43,1.61)
## 5 0.2 (0.11,0.39) 0.66 (0.29,1.46)
## 9 0 (0,Inf) 0 (0,Inf)
##
## masts: ref.=1
## 2 2.3 (1.35,3.91) 1.23 (0.66,2.31)
## 3 2.74 (1.78,4.2) 1.64 (1.01,2.67)
## 4 2.39 (1.13,5.05) 1.12 (0.48,2.64)
## 5 0.87 (0.53,1.43) 0.74 (0.42,1.33)
## 6 1.24 (0.65,2.34) 0.78 (0.38,1.58)
##
## finc (cont. var.) 0.86 (0.82,0.89) 0.91 (0.82,1.02)
##
## PIR (cont. var.) 0.69 (0.62,0.77) 0.92 (0.68,1.23)
##
## SMQ020: ref.=1
## 2 0.59 (0.43,0.81) 0.87 (0.59,1.27)
## 9 0 (0,Inf) 0 (0,Inf)
##
## ALQ101: ref.=1
## 2 1.03 (0.73,1.46) 0.83 (0.55,1.26)
## 9 0 (0,Inf) 0 (0,Inf)
##
## PAQ665: 2 vs 1 2.43 (1.7,3.47) 1.82 (1.24,2.68)
##
## PAQ650: 2 vs 1 3.25 (1.92,5.49) 1.52 (0.85,2.72)
##
## BMI (cont. var.) 1.06 (1.04,1.08) 1.04 (1.02,1.07)
##
## SBP (cont. var.) 1.0086 (1.0003,1.017) 0.9965 (0.9856,1.0075)
##
## DBP (cont. var.) 1.0008 (0.9885,1.0133) 1.0011 (0.9868,1.0156)
##
## AST (cont. var.) 1.0026 (0.9985,1.0066) 1.0017 (0.9947,1.0087)
##
## ALT (cont. var.) 1.005 (0.998,1.0121) 1.0036 (0.9915,1.0157)
##
## BUN (cont. var.) 1.03 (1,1.05) 1.01 (0.98,1.05)
##
## crea (cont. var.) 1.07 (0.85,1.35) 1.13 (0.79,1.62)
##
## TC (cont. var.) 1 (1,1.01) 0.85 (0.47,1.53)
##
## TG (cont. var.) 1.01 (1,1.01) 1.04 (0.92,1.17)
##
## HDL (cont. var.) 0.99 (0.98,1) 1.19 (0.66,2.14)
##
## LDL (cont. var.) 1 (1,1.01) 1.17 (0.65,2.11)
##
## apoB (cont. var.) 1.0091 (1.0029,1.0153) 1.0077 (0.9878,1.028)
##
## P(Wald's test) P(LR-test)
## sCOT (cont. var.) 0.064 0.068
##
## age (cont. var.) 0.494 0.494
##
## sex: 2 vs 1 0.005 0.005
##
## ethnic: ref.=1 0.694
## 2 0.447
## 3 0.656
## 4 0.803
## 6 0.276
## 7 0.896
##
## educ: ref.=1 0.674
## 2 0.62
## 3 0.203
## 4 0.579
## 5 0.305
## 9 0.988
##
## masts: ref.=1 0.2
## 2 0.515
## 3 0.046
## 4 0.793
## 5 0.316
## 6 0.485
##
## finc (cont. var.) 0.109 0.105
##
## PIR (cont. var.) 0.576 0.575
##
## SMQ020: ref.=1 0.677
## 2 0.464
## 9 0.989
##
## ALQ101: ref.=1 0.518
## 2 0.389
## 9 0.978
##
## PAQ665: 2 vs 1 0.002 0.002
##
## PAQ650: 2 vs 1 0.155 0.141
##
## BMI (cont. var.) < 0.001 < 0.001
##
## SBP (cont. var.) 0.53 0.528
##
## DBP (cont. var.) 0.881 0.881
##
## AST (cont. var.) 0.637 0.647
##
## ALT (cont. var.) 0.564 0.57
##
## BUN (cont. var.) 0.412 0.411
##
## crea (cont. var.) 0.493 0.538
##
## TC (cont. var.) 0.586 0.586
##
## TG (cont. var.) 0.536 0.536
##
## HDL (cont. var.) 0.562 0.562
##
## LDL (cont. var.) 0.593 0.593
##
## apoB (cont. var.) 0.452 0.448
##
## Log-likelihood = -498.6712
## No. of observations = 1936
## AIC value = 1075.3423summary(Model4_2)##
## Call:
## glm(formula = dp10 ~ sHCOT + age + sex + ethnic + educ + masts +
## finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 + BMI + SBP +
## DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB,
## family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.5810 -0.4613 -0.3046 -0.1851 3.3808
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -5.375094 1.039564 -5.171 2.33e-07 ***
## sHCOT 0.001424 0.001066 1.336 0.181448
## age 0.004630 0.007459 0.621 0.534834
## sex2 0.575037 0.208958 2.752 0.005925 **
## ethnic2 0.279215 0.352146 0.793 0.427838
## ethnic3 0.153210 0.289383 0.529 0.596503
## ethnic4 0.129321 0.333818 0.387 0.698460
## ethnic6 -0.547393 0.532604 -1.028 0.304060
## ethnic7 0.118463 0.632535 0.187 0.851439
## educ2 -0.162492 0.348120 -0.467 0.640664
## educ3 -0.449882 0.349625 -1.287 0.198180
## educ4 -0.197054 0.338275 -0.583 0.560211
## educ5 -0.443828 0.408570 -1.086 0.277349
## educ9 -13.841853 882.743540 -0.016 0.987489
## masts2 0.205386 0.321185 0.639 0.522521
## masts3 0.498732 0.248551 2.007 0.044797 *
## masts4 0.123695 0.438460 0.282 0.777858
## masts5 -0.276765 0.296408 -0.934 0.350444
## masts6 -0.252265 0.362842 -0.695 0.486901
## finc -0.093045 0.057250 -1.625 0.104114
## PIR -0.080751 0.150415 -0.537 0.591365
## SMQ0202 -0.188241 0.191186 -0.985 0.324823
## SMQ0209 -12.667449 882.743511 -0.014 0.988551
## ALQ1012 -0.184514 0.211820 -0.871 0.383706
## ALQ1019 -12.080990 435.413908 -0.028 0.977865
## PAQ6652 0.601003 0.196609 3.057 0.002237 **
## PAQ6502 0.427161 0.295225 1.447 0.147925
## BMI 0.042822 0.011379 3.763 0.000168 ***
## SBP -0.003325 0.005591 -0.595 0.552047
## DBP 0.001317 0.007370 0.179 0.858197
## AST 0.001585 0.003584 0.442 0.658445
## ALT 0.003533 0.006172 0.572 0.567013
## BUN 0.011936 0.016446 0.726 0.467971
## crea 0.128157 0.183195 0.700 0.484197
## TC -0.156119 0.298962 -0.522 0.601530
## TG 0.035560 0.059760 0.595 0.551809
## HDL 0.165615 0.298874 0.554 0.579491
## LDL 0.153322 0.298711 0.513 0.607757
## apoB 0.007581 0.010143 0.747 0.454812
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1170.32 on 1935 degrees of freedom
## Residual deviance: 998.95 on 1897 degrees of freedom
## (3436 observations deleted due to missingness)
## AIC: 1076.9
##
## Number of Fisher Scoring iterations: 13logistic.display(Model4_2)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## sHCOT (cont. var.) 1.0031 (1.0012,1.005) 1.0014 (0.9993,1.0035)
##
## age (cont. var.) 1.0151 (1.006,1.0242) 1.0046 (0.9901,1.0194)
##
## sex: 2 vs 1 1.75 (1.26,2.41) 1.78 (1.18,2.68)
##
## ethnic: ref.=1
## 2 1.16 (0.62,2.2) 1.32 (0.66,2.64)
## 3 1 (0.62,1.61) 1.17 (0.66,2.06)
## 4 0.89 (0.51,1.55) 1.14 (0.59,2.19)
## 6 0.22 (0.08,0.59) 0.58 (0.2,1.64)
## 7 0.89 (0.29,2.7) 1.13 (0.33,3.89)
##
## educ: ref.=1
## 2 0.66 (0.36,1.19) 0.85 (0.43,1.68)
## 3 0.45 (0.25,0.81) 0.64 (0.32,1.27)
## 4 0.52 (0.31,0.9) 0.82 (0.42,1.59)
## 5 0.2 (0.11,0.39) 0.64 (0.29,1.43)
## 9 0 (0,Inf) 0 (0,Inf)
##
## masts: ref.=1
## 2 2.3 (1.35,3.91) 1.23 (0.65,2.3)
## 3 2.74 (1.78,4.2) 1.65 (1.01,2.68)
## 4 2.39 (1.13,5.05) 1.13 (0.48,2.67)
## 5 0.87 (0.53,1.43) 0.76 (0.42,1.36)
## 6 1.24 (0.65,2.34) 0.78 (0.38,1.58)
##
## finc (cont. var.) 0.86 (0.82,0.89) 0.91 (0.81,1.02)
##
## PIR (cont. var.) 0.69 (0.62,0.77) 0.92 (0.69,1.24)
##
## SMQ020: ref.=1
## 2 0.59 (0.43,0.81) 0.83 (0.57,1.2)
## 9 0 (0,Inf) 0 (0,Inf)
##
## ALQ101: ref.=1
## 2 1.03 (0.73,1.46) 0.83 (0.55,1.26)
## 9 0 (0,Inf) 0 (0,Inf)
##
## PAQ665: 2 vs 1 2.43 (1.7,3.47) 1.82 (1.24,2.68)
##
## PAQ650: 2 vs 1 3.25 (1.92,5.49) 1.53 (0.86,2.73)
##
## BMI (cont. var.) 1.06 (1.04,1.08) 1.04 (1.02,1.07)
##
## SBP (cont. var.) 1.0086 (1.0003,1.017) 0.9967 (0.9858,1.0077)
##
## DBP (cont. var.) 1.0008 (0.9885,1.0133) 1.0013 (0.987,1.0159)
##
## AST (cont. var.) 1.0026 (0.9985,1.0066) 1.0016 (0.9946,1.0086)
##
## ALT (cont. var.) 1.005 (0.998,1.0121) 1.0035 (0.9915,1.0158)
##
## BUN (cont. var.) 1.03 (1,1.05) 1.01 (0.98,1.05)
##
## crea (cont. var.) 1.07 (0.85,1.35) 1.14 (0.79,1.63)
##
## TC (cont. var.) 1 (1,1.01) 0.86 (0.48,1.54)
##
## TG (cont. var.) 1.01 (1,1.01) 1.04 (0.92,1.16)
##
## HDL (cont. var.) 0.99 (0.98,1) 1.18 (0.66,2.12)
##
## LDL (cont. var.) 1 (1,1.01) 1.17 (0.65,2.09)
##
## apoB (cont. var.) 1.0091 (1.0029,1.0153) 1.0076 (0.9878,1.0278)
##
## P(Wald's test) P(LR-test)
## sHCOT (cont. var.) 0.181 0.187
##
## age (cont. var.) 0.535 0.535
##
## sex: 2 vs 1 0.006 0.006
##
## ethnic: ref.=1 0.705
## 2 0.428
## 3 0.597
## 4 0.698
## 6 0.304
## 7 0.851
##
## educ: ref.=1 0.65
## 2 0.641
## 3 0.198
## 4 0.56
## 5 0.277
## 9 0.987
##
## masts: ref.=1 0.214
## 2 0.523
## 3 0.045
## 4 0.778
## 5 0.35
## 6 0.487
##
## finc (cont. var.) 0.104 0.101
##
## PIR (cont. var.) 0.591 0.591
##
## SMQ020: ref.=1 0.539
## 2 0.325
## 9 0.989
##
## ALQ101: ref.=1 0.504
## 2 0.384
## 9 0.978
##
## PAQ665: 2 vs 1 0.002 0.002
##
## PAQ650: 2 vs 1 0.148 0.135
##
## BMI (cont. var.) < 0.001 < 0.001
##
## SBP (cont. var.) 0.552 0.55
##
## DBP (cont. var.) 0.858 0.858
##
## AST (cont. var.) 0.658 0.668
##
## ALT (cont. var.) 0.567 0.572
##
## BUN (cont. var.) 0.468 0.467
##
## crea (cont. var.) 0.484 0.53
##
## TC (cont. var.) 0.602 0.601
##
## TG (cont. var.) 0.552 0.552
##
## HDL (cont. var.) 0.579 0.579
##
## LDL (cont. var.) 0.608 0.608
##
## apoB (cont. var.) 0.455 0.451
##
## Log-likelihood = -499.4738
## No. of observations = 1936
## AIC value = 1076.9475summary(Model4_3)##
## Call:
## glm(formula = dp10 ~ uCOT + age + sex + ethnic + educ + masts +
## finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 + BMI + SBP +
## DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB,
## family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.7043 -0.3850 -0.2423 -0.1348 2.9197
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -4.111e+00 2.291e+00 -1.794 0.0727 .
## uCOT 1.357e-04 6.405e-05 2.118 0.0342 *
## age 1.514e-02 1.490e-02 1.016 0.3095
## sex2 5.587e-01 4.462e-01 1.252 0.2105
## ethnic2 6.054e-01 8.631e-01 0.701 0.4830
## ethnic3 1.307e+00 7.473e-01 1.749 0.0803 .
## ethnic4 1.259e+00 8.370e-01 1.504 0.1325
## ethnic6 1.251e+00 1.060e+00 1.180 0.2379
## ethnic7 -1.281e+01 7.069e+02 -0.018 0.9855
## educ2 -6.786e-01 7.617e-01 -0.891 0.3730
## educ3 -8.407e-01 7.087e-01 -1.186 0.2355
## educ4 -2.803e-01 6.836e-01 -0.410 0.6817
## educ5 -7.310e-01 8.044e-01 -0.909 0.3634
## masts2 -5.876e-03 7.147e-01 -0.008 0.9934
## masts3 7.187e-01 4.902e-01 1.466 0.1426
## masts4 5.946e-01 8.098e-01 0.734 0.4628
## masts5 -6.124e-01 6.052e-01 -1.012 0.3116
## masts6 -5.664e-01 7.675e-01 -0.738 0.4606
## finc -1.922e-01 1.233e-01 -1.559 0.1190
## PIR -2.591e-02 3.168e-01 -0.082 0.9348
## SMQ0202 1.673e-02 3.926e-01 0.043 0.9660
## ALQ1012 -8.142e-01 4.646e-01 -1.753 0.0797 .
## PAQ6652 3.973e-01 3.908e-01 1.017 0.3093
## PAQ6502 4.231e-01 5.606e-01 0.755 0.4503
## BMI 3.988e-02 2.626e-02 1.519 0.1289
## SBP -2.914e-02 1.314e-02 -2.219 0.0265 *
## DBP 1.441e-02 1.594e-02 0.904 0.3659
## AST 2.486e-02 2.365e-02 1.051 0.2932
## ALT -8.692e-03 2.152e-02 -0.404 0.6862
## BUN 9.009e-03 3.525e-02 0.256 0.7983
## crea 5.627e-01 5.442e-01 1.034 0.3012
## TC -8.099e-01 6.261e-01 -1.294 0.1958
## TG 1.632e-01 1.252e-01 1.303 0.1924
## HDL 8.167e-01 6.259e-01 1.305 0.1919
## LDL 8.194e-01 6.263e-01 1.308 0.1908
## apoB -5.891e-03 2.129e-02 -0.277 0.7820
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 325.77 on 606 degrees of freedom
## Residual deviance: 256.84 on 571 degrees of freedom
## (4765 observations deleted due to missingness)
## AIC: 328.84
##
## Number of Fisher Scoring iterations: 15logistic.display(Model4_3)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## uCOT (cont. var.) 1.0002 (1.0001,1.0003) 1.0001 (1,1.0003)
##
## age (cont. var.) 1.02 (1,1.03) 1.02 (0.99,1.05)
##
## sex: 2 vs 1 1.4 (0.76,2.57) 1.75 (0.73,4.19)
##
## ethnic: ref.=1
## 2 2.01 (0.43,9.33) 1.83 (0.34,9.94)
## 3 3.16 (0.94,10.66) 3.69 (0.85,15.99)
## 4 2.61 (0.68,9.93) 3.52 (0.68,18.17)
## 6 1.04 (0.17,6.44) 3.49 (0.44,27.86)
## 7 0 (0,Inf) 0 (0,Inf)
##
## educ: ref.=1
## 2 0.62 (0.18,2.16) 0.51 (0.11,2.26)
## 3 0.65 (0.21,2.01) 0.43 (0.11,1.73)
## 4 0.72 (0.25,2.05) 0.76 (0.2,2.88)
## 5 0.28 (0.08,0.98) 0.48 (0.1,2.33)
##
## masts: ref.=1
## 2 2.4856 (0.7849,7.8714) 0.9941 (0.2449,4.0348)
## 3 3.87 (1.76,8.51) 2.05 (0.79,5.36)
## 4 3.73 (1.14,12.17) 1.81 (0.37,8.86)
## 5 0.7 (0.26,1.94) 0.54 (0.17,1.77)
## 6 1.17 (0.33,4.14) 0.57 (0.13,2.55)
##
## finc (cont. var.) 0.83 (0.75,0.91) 0.83 (0.65,1.05)
##
## PIR (cont. var.) 0.65 (0.52,0.82) 0.97 (0.52,1.81)
##
## SMQ020: 2 vs 1 0.54 (0.29,1) 1.02 (0.47,2.2)
##
## ALQ101: 2 vs 1 0.89 (0.44,1.79) 0.44 (0.18,1.1)
##
## PAQ665: 2 vs 1 1.72 (0.9,3.3) 1.49 (0.69,3.2)
##
## PAQ650: 2 vs 1 2.55 (0.99,6.58) 1.53 (0.51,4.58)
##
## BMI (cont. var.) 1.03 (0.99,1.07) 1.04 (0.99,1.1)
##
## SBP (cont. var.) 0.99 (0.98,1.01) 0.97 (0.95,1)
##
## DBP (cont. var.) 1 (0.98,1.03) 1.01 (0.98,1.05)
##
## AST (cont. var.) 1.01 (0.99,1.04) 1.03 (0.98,1.07)
##
## ALT (cont. var.) 1.0005 (0.9782,1.0234) 0.9913 (0.9504,1.034)
##
## BUN (cont. var.) 1.039 (0.9945,1.0854) 1.009 (0.9417,1.0812)
##
## crea (cont. var.) 1.92 (0.95,3.85) 1.76 (0.6,5.1)
##
## TC (cont. var.) 1 (1,1.01) 0.44 (0.13,1.52)
##
## TG (cont. var.) 1 (1,1.01) 1.18 (0.92,1.5)
##
## HDL (cont. var.) 0.99 (0.97,1.01) 2.26 (0.66,7.72)
##
## LDL (cont. var.) 1 (1,1.01) 2.27 (0.66,7.74)
##
## apoB (cont. var.) 1.0052 (0.9936,1.017) 0.9941 (0.9535,1.0365)
##
## P(Wald's test) P(LR-test)
## uCOT (cont. var.) 0.034 0.029
##
## age (cont. var.) 0.31 0.309
##
## sex: 2 vs 1 0.21 0.209
##
## ethnic: ref.=1 0.347
## 2 0.483
## 3 0.08
## 4 0.132
## 6 0.238
## 7 0.986
##
## educ: ref.=1 0.637
## 2 0.373
## 3 0.236
## 4 0.682
## 5 0.363
##
## masts: ref.=1 0.363
## 2 0.993
## 3 0.143
## 4 0.463
## 5 0.312
## 6 0.461
##
## finc (cont. var.) 0.119 0.108
##
## PIR (cont. var.) 0.935 0.935
##
## SMQ020: 2 vs 1 0.966 0.966
##
## ALQ101: 2 vs 1 0.08 0.069
##
## PAQ665: 2 vs 1 0.309 0.304
##
## PAQ650: 2 vs 1 0.45 0.438
##
## BMI (cont. var.) 0.129 0.133
##
## SBP (cont. var.) 0.027 0.02
##
## DBP (cont. var.) 0.366 0.351
##
## AST (cont. var.) 0.293 0.309
##
## ALT (cont. var.) 0.686 0.682
##
## BUN (cont. var.) 0.798 0.799
##
## crea (cont. var.) 0.301 0.306
##
## TC (cont. var.) 0.196 0.191
##
## TG (cont. var.) 0.192 0.188
##
## HDL (cont. var.) 0.192 0.188
##
## LDL (cont. var.) 0.191 0.186
##
## apoB (cont. var.) 0.782 0.783
##
## Log-likelihood = -128.4201
## No. of observations = 607
## AIC value = 328.8401summary(Model4_4)##
## Call:
## glm(formula = dp10 ~ uHCOT + age + sex + ethnic + educ + masts +
## finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 + BMI + SBP +
## DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB,
## family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.4144 -0.3897 -0.2435 -0.1343 2.8941
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -3.769e+00 2.271e+00 -1.660 0.0969 .
## uHCOT 3.053e-05 2.208e-05 1.383 0.1668
## age 1.177e-02 1.467e-02 0.802 0.4226
## sex2 4.747e-01 4.415e-01 1.075 0.2823
## ethnic2 6.262e-01 8.587e-01 0.729 0.4658
## ethnic3 1.393e+00 7.468e-01 1.865 0.0622 .
## ethnic4 1.315e+00 8.360e-01 1.573 0.1156
## ethnic6 1.239e+00 1.059e+00 1.170 0.2418
## ethnic7 -1.274e+01 7.084e+02 -0.018 0.9856
## educ2 -6.574e-01 7.621e-01 -0.863 0.3883
## educ3 -7.260e-01 6.986e-01 -1.039 0.2987
## educ4 -2.717e-01 6.849e-01 -0.397 0.6916
## educ5 -7.280e-01 8.068e-01 -0.902 0.3669
## masts2 1.876e-02 7.154e-01 0.026 0.9791
## masts3 7.711e-01 4.856e-01 1.588 0.1123
## masts4 6.784e-01 8.061e-01 0.842 0.4000
## masts5 -6.195e-01 6.053e-01 -1.023 0.3061
## masts6 -5.244e-01 7.645e-01 -0.686 0.4928
## finc -1.832e-01 1.229e-01 -1.491 0.1359
## PIR -4.844e-02 3.154e-01 -0.154 0.8780
## SMQ0202 -7.683e-02 3.856e-01 -0.199 0.8421
## ALQ1012 -7.070e-01 4.529e-01 -1.561 0.1186
## PAQ6652 3.888e-01 3.879e-01 1.002 0.3163
## PAQ6502 4.899e-01 5.587e-01 0.877 0.3806
## BMI 3.653e-02 2.613e-02 1.398 0.1622
## SBP -2.924e-02 1.316e-02 -2.223 0.0262 *
## DBP 1.508e-02 1.607e-02 0.939 0.3479
## AST 2.510e-02 2.352e-02 1.067 0.2860
## ALT -9.660e-03 2.159e-02 -0.447 0.6546
## BUN 1.248e-02 3.502e-02 0.356 0.7215
## crea 4.811e-01 5.394e-01 0.892 0.3724
## TC -9.449e-01 6.182e-01 -1.528 0.1264
## TG 1.899e-01 1.236e-01 1.536 0.1246
## HDL 9.496e-01 6.181e-01 1.536 0.1245
## LDL 9.555e-01 6.184e-01 1.545 0.1223
## apoB -6.665e-03 2.110e-02 -0.316 0.7521
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 325.77 on 606 degrees of freedom
## Residual deviance: 259.85 on 571 degrees of freedom
## (4765 observations deleted due to missingness)
## AIC: 331.85
##
## Number of Fisher Scoring iterations: 15logistic.display(Model4_4)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## uHCOT (cont. var.) 1 (1,1.0001) 1 (1,1.0001)
##
## age (cont. var.) 1.02 (1,1.03) 1.01 (0.98,1.04)
##
## sex: 2 vs 1 1.4 (0.76,2.57) 1.61 (0.68,3.82)
##
## ethnic: ref.=1
## 2 2.01 (0.43,9.33) 1.87 (0.35,10.07)
## 3 3.16 (0.94,10.66) 4.03 (0.93,17.4)
## 4 2.61 (0.68,9.93) 3.73 (0.72,19.18)
## 6 1.04 (0.17,6.44) 3.45 (0.43,27.5)
## 7 0 (0,Inf) 0 (0,Inf)
##
## educ: ref.=1
## 2 0.62 (0.18,2.16) 0.52 (0.12,2.31)
## 3 0.65 (0.21,2.01) 0.48 (0.12,1.9)
## 4 0.72 (0.25,2.05) 0.76 (0.2,2.92)
## 5 0.28 (0.08,0.98) 0.48 (0.1,2.35)
##
## masts: ref.=1
## 2 2.49 (0.78,7.87) 1.02 (0.25,4.14)
## 3 3.87 (1.76,8.51) 2.16 (0.83,5.6)
## 4 3.73 (1.14,12.17) 1.97 (0.41,9.57)
## 5 0.7 (0.26,1.94) 0.54 (0.16,1.76)
## 6 1.17 (0.33,4.14) 0.59 (0.13,2.65)
##
## finc (cont. var.) 0.83 (0.75,0.91) 0.83 (0.65,1.06)
##
## PIR (cont. var.) 0.65 (0.52,0.82) 0.95 (0.51,1.77)
##
## SMQ020: 2 vs 1 0.54 (0.29,1) 0.93 (0.43,1.97)
##
## ALQ101: 2 vs 1 0.89 (0.44,1.79) 0.49 (0.2,1.2)
##
## PAQ665: 2 vs 1 1.72 (0.9,3.3) 1.48 (0.69,3.16)
##
## PAQ650: 2 vs 1 2.55 (0.99,6.58) 1.63 (0.55,4.88)
##
## BMI (cont. var.) 1.03 (0.99,1.07) 1.04 (0.99,1.09)
##
## SBP (cont. var.) 0.99 (0.98,1.01) 0.97 (0.95,1)
##
## DBP (cont. var.) 1 (0.98,1.03) 1.02 (0.98,1.05)
##
## AST (cont. var.) 1.01 (0.99,1.04) 1.03 (0.98,1.07)
##
## ALT (cont. var.) 1.0005 (0.9782,1.0234) 0.9904 (0.9493,1.0332)
##
## BUN (cont. var.) 1.04 (0.99,1.09) 1.01 (0.95,1.08)
##
## crea (cont. var.) 1.92 (0.95,3.85) 1.62 (0.56,4.66)
##
## TC (cont. var.) 1 (1,1.01) 0.39 (0.12,1.31)
##
## TG (cont. var.) 1 (1,1.01) 1.21 (0.95,1.54)
##
## HDL (cont. var.) 0.99 (0.97,1.01) 2.58 (0.77,8.68)
##
## LDL (cont. var.) 1 (1,1.01) 2.6 (0.77,8.74)
##
## apoB (cont. var.) 1.0052 (0.9936,1.017) 0.9934 (0.9531,1.0353)
##
## P(Wald's test) P(LR-test)
## uHCOT (cont. var.) 0.167 0.183
##
## age (cont. var.) 0.423 0.423
##
## sex: 2 vs 1 0.282 0.281
##
## ethnic: ref.=1 0.293
## 2 0.466
## 3 0.062
## 4 0.116
## 6 0.242
## 7 0.986
##
## educ: ref.=1 0.719
## 2 0.388
## 3 0.299
## 4 0.692
## 5 0.367
##
## masts: ref.=1 0.31
## 2 0.979
## 3 0.112
## 4 0.4
## 5 0.306
## 6 0.493
##
## finc (cont. var.) 0.136 0.125
##
## PIR (cont. var.) 0.878 0.878
##
## SMQ020: 2 vs 1 0.842 0.842
##
## ALQ101: 2 vs 1 0.119 0.108
##
## PAQ665: 2 vs 1 0.316 0.311
##
## PAQ650: 2 vs 1 0.381 0.365
##
## BMI (cont. var.) 0.162 0.166
##
## SBP (cont. var.) 0.026 0.02
##
## DBP (cont. var.) 0.348 0.332
##
## AST (cont. var.) 0.286 0.302
##
## ALT (cont. var.) 0.655 0.649
##
## BUN (cont. var.) 0.722 0.723
##
## crea (cont. var.) 0.372 0.377
##
## TC (cont. var.) 0.126 0.122
##
## TG (cont. var.) 0.125 0.12
##
## HDL (cont. var.) 0.124 0.12
##
## LDL (cont. var.) 0.122 0.118
##
## apoB (cont. var.) 0.752 0.754
##
## Log-likelihood = -129.9242
## No. of observations = 607
## AIC value = 331.84838.6 Model 5
### Adjust for: age, sex, ethnic group, educ, marrital status + finc + PIR + smoking + alcohol + physical activity + BMI + SBP + DBP + AST + ALT + BUN + Crea + TC + TG + HDL + LDL + ApoB + hyper + diab + dyslip
Model5_1 = glm(dp10 ~ sCOT + age + sex + ethnic + educ + masts + finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 + BMI + SBP + DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB + hyper + db + dyslip, family = binomial, data = tab1)
Model5_2 = glm(dp10 ~ sHCOT + age + sex + ethnic + educ + masts + finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 +BMI + SBP + DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB + hyper + db + dyslip, family = binomial, data = tab1)
Model5_3 = glm(dp10 ~ uCOT + age + sex + ethnic + educ + masts + finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 + BMI + SBP + DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB + hyper + db + dyslip, family = binomial, data = tab1)
Model5_4 = glm(dp10 ~ uHCOT + age + sex + ethnic + educ + masts + finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 + BMI + SBP + DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB + hyper + db + dyslip, family = binomial, data = tab1)
# Summary the results of logistic regression
summary(Model5_1)##
## Call:
## glm(formula = dp10 ~ sCOT + age + sex + ethnic + educ + masts +
## finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 + BMI + SBP +
## DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB +
## hyper + db + dyslip, family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.4731 -0.4645 -0.3015 -0.1799 3.4276
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -4.700e+00 1.103e+00 -4.260 2.04e-05 ***
## sCOT 1.010e-03 5.661e-04 1.784 0.074492 .
## age -8.583e-05 8.146e-03 -0.011 0.991594
## sex2 5.667e-01 2.107e-01 2.690 0.007150 **
## ethnic2 2.774e-01 3.536e-01 0.785 0.432709
## ethnic3 1.196e-01 2.893e-01 0.413 0.679302
## ethnic4 2.893e-02 3.368e-01 0.086 0.931539
## ethnic6 -6.089e-01 5.338e-01 -1.141 0.254034
## ethnic7 9.657e-02 6.241e-01 0.155 0.877035
## educ2 -2.001e-01 3.487e-01 -0.574 0.565944
## educ3 -5.077e-01 3.511e-01 -1.446 0.148108
## educ4 -2.371e-01 3.384e-01 -0.701 0.483456
## educ5 -4.483e-01 4.089e-01 -1.096 0.272921
## educ9 -1.483e+01 1.455e+03 -0.010 0.991869
## masts2 2.103e-01 3.219e-01 0.653 0.513452
## masts3 5.003e-01 2.495e-01 2.005 0.044915 *
## masts4 9.218e-02 4.385e-01 0.210 0.833507
## masts5 -3.002e-01 2.968e-01 -1.012 0.311654
## masts6 -2.638e-01 3.641e-01 -0.725 0.468708
## finc -9.393e-02 5.738e-02 -1.637 0.101660
## PIR -7.574e-02 1.503e-01 -0.504 0.614277
## SMQ0202 -1.345e-01 1.960e-01 -0.686 0.492477
## SMQ0209 -1.315e+01 1.455e+03 -0.009 0.992794
## ALQ1012 -1.881e-01 2.130e-01 -0.883 0.377302
## ALQ1019 -1.300e+01 7.133e+02 -0.018 0.985458
## PAQ6652 5.698e-01 1.976e-01 2.884 0.003930 **
## PAQ6502 4.366e-01 2.960e-01 1.475 0.140158
## BMI 3.998e-02 1.171e-02 3.415 0.000637 ***
## SBP -7.295e-03 5.875e-03 -1.242 0.214319
## DBP 1.651e-04 7.360e-03 0.022 0.982099
## AST 9.757e-04 3.618e-03 0.270 0.787440
## ALT 3.905e-03 6.221e-03 0.628 0.530167
## BUN 1.105e-02 1.657e-02 0.667 0.504634
## crea 1.159e-01 1.812e-01 0.640 0.522395
## TC -1.756e-01 2.998e-01 -0.586 0.558129
## TG 3.956e-02 5.992e-02 0.660 0.509098
## HDL 1.861e-01 2.998e-01 0.621 0.534724
## LDL 1.740e-01 2.995e-01 0.581 0.561385
## apoB 7.840e-03 1.025e-02 0.765 0.444180
## hyper1 4.334e-01 2.223e-01 1.950 0.051225 .
## db1 1.497e-01 2.134e-01 0.702 0.482941
## dyslip1 -2.129e-01 2.827e-01 -0.753 0.451290
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1170.32 on 1935 degrees of freedom
## Residual deviance: 992.63 on 1894 degrees of freedom
## (3436 observations deleted due to missingness)
## AIC: 1076.6
##
## Number of Fisher Scoring iterations: 14logistic.display(Model5_1)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## sCOT (cont. var.) 1.0015 (1.0006,1.0024) 1.001 (0.9999,1.0021)
##
## age (cont. var.) 1.0151 (1.006,1.0242) 0.9999 (0.9841,1.016)
##
## sex: 2 vs 1 1.75 (1.26,2.41) 1.76 (1.17,2.66)
##
## ethnic: ref.=1
## 2 1.16 (0.62,2.2) 1.32 (0.66,2.64)
## 3 1 (0.62,1.61) 1.13 (0.64,1.99)
## 4 0.89 (0.51,1.55) 1.03 (0.53,1.99)
## 6 0.22 (0.08,0.59) 0.54 (0.19,1.55)
## 7 0.89 (0.29,2.7) 1.1 (0.32,3.74)
##
## educ: ref.=1
## 2 0.66 (0.36,1.19) 0.82 (0.41,1.62)
## 3 0.45 (0.25,0.81) 0.6 (0.3,1.2)
## 4 0.52 (0.31,0.9) 0.79 (0.41,1.53)
## 5 0.2 (0.11,0.39) 0.64 (0.29,1.42)
## 9 0 (0,Inf) 0 (0,Inf)
##
## masts: ref.=1
## 2 2.3 (1.35,3.91) 1.23 (0.66,2.32)
## 3 2.74 (1.78,4.2) 1.65 (1.01,2.69)
## 4 2.39 (1.13,5.05) 1.1 (0.46,2.59)
## 5 0.87 (0.53,1.43) 0.74 (0.41,1.32)
## 6 1.24 (0.65,2.34) 0.77 (0.38,1.57)
##
## finc (cont. var.) 0.86 (0.82,0.89) 0.91 (0.81,1.02)
##
## PIR (cont. var.) 0.69 (0.62,0.77) 0.93 (0.69,1.24)
##
## SMQ020: ref.=1
## 2 0.59 (0.43,0.81) 0.87 (0.6,1.28)
## 9 0 (0,Inf) 0 (0,Inf)
##
## ALQ101: ref.=1
## 2 1.03 (0.73,1.46) 0.83 (0.55,1.26)
## 9 0 (0,Inf) 0 (0,Inf)
##
## PAQ665: 2 vs 1 2.43 (1.7,3.47) 1.77 (1.2,2.6)
##
## PAQ650: 2 vs 1 3.25 (1.92,5.49) 1.55 (0.87,2.76)
##
## BMI (cont. var.) 1.06 (1.04,1.08) 1.04 (1.02,1.06)
##
## SBP (cont. var.) 1.0086 (1.0003,1.017) 0.9927 (0.9814,1.0042)
##
## DBP (cont. var.) 1.0008 (0.9885,1.0133) 1.0002 (0.9858,1.0147)
##
## AST (cont. var.) 1.0026 (0.9985,1.0066) 1.001 (0.9939,1.0081)
##
## ALT (cont. var.) 1.005 (0.998,1.0121) 1.0039 (0.9917,1.0162)
##
## BUN (cont. var.) 1.03 (1,1.05) 1.01 (0.98,1.04)
##
## crea (cont. var.) 1.07 (0.85,1.35) 1.12 (0.79,1.6)
##
## TC (cont. var.) 1 (1,1.01) 0.84 (0.47,1.51)
##
## TG (cont. var.) 1.01 (1,1.01) 1.04 (0.93,1.17)
##
## HDL (cont. var.) 0.99 (0.98,1) 1.2 (0.67,2.17)
##
## LDL (cont. var.) 1 (1,1.01) 1.19 (0.66,2.14)
##
## apoB (cont. var.) 1.0091 (1.0029,1.0153) 1.0079 (0.9878,1.0283)
##
## hyper: 1 vs 0 2.23 (1.63,3.05) 1.54 (1,2.38)
##
## db: 1 vs 0 1.89 (1.33,2.69) 1.16 (0.76,1.76)
##
## dyslip: 1 vs 0 1.53 (0.99,2.35) 0.81 (0.46,1.41)
##
## P(Wald's test) P(LR-test)
## sCOT (cont. var.) 0.074 0.078
##
## age (cont. var.) 0.992 0.992
##
## sex: 2 vs 1 0.007 0.007
##
## ethnic: ref.=1 0.654
## 2 0.433
## 3 0.679
## 4 0.932
## 6 0.254
## 7 0.877
##
## educ: ref.=1 0.613
## 2 0.566
## 3 0.148
## 4 0.483
## 5 0.273
## 9 0.992
##
## masts: ref.=1 0.194
## 2 0.513
## 3 0.045
## 4 0.834
## 5 0.312
## 6 0.469
##
## finc (cont. var.) 0.102 0.098
##
## PIR (cont. var.) 0.614 0.614
##
## SMQ020: ref.=1 0.729
## 2 0.492
## 9 0.993
##
## ALQ101: ref.=1 0.508
## 2 0.377
## 9 0.985
##
## PAQ665: 2 vs 1 0.004 0.003
##
## PAQ650: 2 vs 1 0.14 0.127
##
## BMI (cont. var.) < 0.001 < 0.001
##
## SBP (cont. var.) 0.214 0.211
##
## DBP (cont. var.) 0.982 0.982
##
## AST (cont. var.) 0.787 0.791
##
## ALT (cont. var.) 0.53 0.537
##
## BUN (cont. var.) 0.505 0.504
##
## crea (cont. var.) 0.522 0.563
##
## TC (cont. var.) 0.558 0.558
##
## TG (cont. var.) 0.509 0.509
##
## HDL (cont. var.) 0.535 0.535
##
## LDL (cont. var.) 0.561 0.561
##
## apoB (cont. var.) 0.444 0.44
##
## hyper: 1 vs 0 0.051 0.051
##
## db: 1 vs 0 0.483 0.485
##
## dyslip: 1 vs 0 0.451 0.454
##
## Log-likelihood = -496.313
## No. of observations = 1936
## AIC value = 1076.6261summary(Model5_2)##
## Call:
## glm(formula = dp10 ~ sHCOT + age + sex + ethnic + educ + masts +
## finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 + BMI + SBP +
## DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB +
## hyper + db + dyslip, family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.4425 -0.4629 -0.3010 -0.1798 3.4143
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -4.587e+00 1.100e+00 -4.169 3.06e-05 ***
## sHCOT 1.333e-03 1.073e-03 1.242 0.214282
## age -5.302e-04 8.125e-03 -0.065 0.947970
## sex2 5.540e-01 2.101e-01 2.636 0.008379 **
## ethnic2 2.884e-01 3.532e-01 0.817 0.414154
## ethnic3 1.448e-01 2.895e-01 0.500 0.617031
## ethnic4 7.489e-02 3.342e-01 0.224 0.822707
## ethnic6 -5.773e-01 5.330e-01 -1.083 0.278778
## ethnic7 1.338e-01 6.245e-01 0.214 0.830323
## educ2 -1.899e-01 3.486e-01 -0.545 0.585944
## educ3 -5.131e-01 3.511e-01 -1.461 0.143909
## educ4 -2.470e-01 3.388e-01 -0.729 0.465984
## educ5 -4.728e-01 4.087e-01 -1.157 0.247351
## educ9 -1.485e+01 1.455e+03 -0.010 0.991861
## masts2 2.067e-01 3.219e-01 0.642 0.520835
## masts3 5.027e-01 2.495e-01 2.015 0.043916 *
## masts4 1.017e-01 4.400e-01 0.231 0.817232
## masts5 -2.806e-01 2.963e-01 -0.947 0.343702
## masts6 -2.633e-01 3.640e-01 -0.723 0.469484
## finc -9.521e-02 5.739e-02 -1.659 0.097127 .
## PIR -7.258e-02 1.504e-01 -0.483 0.629365
## SMQ0202 -1.807e-01 1.922e-01 -0.940 0.347024
## SMQ0209 -1.323e+01 1.455e+03 -0.009 0.992749
## ALQ1012 -1.904e-01 2.128e-01 -0.895 0.370931
## ALQ1019 -1.307e+01 7.119e+02 -0.018 0.985354
## PAQ6652 5.716e-01 1.975e-01 2.895 0.003794 **
## PAQ6502 4.440e-01 2.960e-01 1.500 0.133558
## BMI 3.942e-02 1.170e-02 3.368 0.000757 ***
## SBP -7.124e-03 5.881e-03 -1.211 0.225717
## DBP 3.943e-04 7.369e-03 0.054 0.957326
## AST 8.742e-04 3.627e-03 0.241 0.809535
## ALT 3.897e-03 6.236e-03 0.625 0.532068
## BUN 9.517e-03 1.654e-02 0.575 0.564983
## crea 1.187e-01 1.810e-01 0.656 0.511817
## TC -1.693e-01 3.000e-01 -0.564 0.572484
## TG 3.820e-02 5.995e-02 0.637 0.523990
## HDL 1.789e-01 2.999e-01 0.596 0.550922
## LDL 1.679e-01 2.997e-01 0.560 0.575208
## apoB 7.761e-03 1.021e-02 0.760 0.447314
## hyper1 4.318e-01 2.225e-01 1.940 0.052327 .
## db1 1.505e-01 2.132e-01 0.706 0.480436
## dyslip1 -2.172e-01 2.824e-01 -0.769 0.441773
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1170.32 on 1935 degrees of freedom
## Residual deviance: 994.23 on 1894 degrees of freedom
## (3436 observations deleted due to missingness)
## AIC: 1078.2
##
## Number of Fisher Scoring iterations: 14logistic.display(Model5_2)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## sHCOT (cont. var.) 1.0031 (1.0012,1.005) 1.0013 (0.9992,1.0034)
##
## age (cont. var.) 1.0151 (1.006,1.0242) 0.9995 (0.9837,1.0155)
##
## sex: 2 vs 1 1.75 (1.26,2.41) 1.74 (1.15,2.63)
##
## ethnic: ref.=1
## 2 1.16 (0.62,2.2) 1.33 (0.67,2.67)
## 3 1 (0.62,1.61) 1.16 (0.66,2.04)
## 4 0.89 (0.51,1.55) 1.08 (0.56,2.07)
## 6 0.22 (0.08,0.59) 0.56 (0.2,1.6)
## 7 0.89 (0.29,2.7) 1.14 (0.34,3.89)
##
## educ: ref.=1
## 2 0.66 (0.36,1.19) 0.83 (0.42,1.64)
## 3 0.45 (0.25,0.81) 0.6 (0.3,1.19)
## 4 0.52 (0.31,0.9) 0.78 (0.4,1.52)
## 5 0.2 (0.11,0.39) 0.62 (0.28,1.39)
## 9 0 (0,Inf) 0 (0,Inf)
##
## masts: ref.=1
## 2 2.3 (1.35,3.91) 1.23 (0.65,2.31)
## 3 2.74 (1.78,4.2) 1.65 (1.01,2.7)
## 4 2.39 (1.13,5.05) 1.11 (0.47,2.62)
## 5 0.87 (0.53,1.43) 0.76 (0.42,1.35)
## 6 1.24 (0.65,2.34) 0.77 (0.38,1.57)
##
## finc (cont. var.) 0.86 (0.82,0.89) 0.91 (0.81,1.02)
##
## PIR (cont. var.) 0.69 (0.62,0.77) 0.93 (0.69,1.25)
##
## SMQ020: ref.=1
## 2 0.59 (0.43,0.81) 0.83 (0.57,1.22)
## 9 0 (0,Inf) 0 (0,Inf)
##
## ALQ101: ref.=1
## 2 1.03 (0.73,1.46) 0.83 (0.54,1.25)
## 9 0 (0,Inf) 0 (0,Inf)
##
## PAQ665: 2 vs 1 2.43 (1.7,3.47) 1.77 (1.2,2.61)
##
## PAQ650: 2 vs 1 3.25 (1.92,5.49) 1.56 (0.87,2.78)
##
## BMI (cont. var.) 1.06 (1.04,1.08) 1.04 (1.02,1.06)
##
## SBP (cont. var.) 1.0086 (1.0003,1.017) 0.9929 (0.9815,1.0044)
##
## DBP (cont. var.) 1.0008 (0.9885,1.0133) 1.0004 (0.986,1.0149)
##
## AST (cont. var.) 1.0026 (0.9985,1.0066) 1.0009 (0.9938,1.008)
##
## ALT (cont. var.) 1.005 (0.998,1.0121) 1.0039 (0.9917,1.0162)
##
## BUN (cont. var.) 1.0252 (1.0014,1.0495) 1.0096 (0.9774,1.0428)
##
## crea (cont. var.) 1.07 (0.85,1.35) 1.13 (0.79,1.61)
##
## TC (cont. var.) 1 (1,1.01) 0.84 (0.47,1.52)
##
## TG (cont. var.) 1.01 (1,1.01) 1.04 (0.92,1.17)
##
## HDL (cont. var.) 0.99 (0.98,1) 1.2 (0.66,2.15)
##
## LDL (cont. var.) 1 (1,1.01) 1.18 (0.66,2.13)
##
## apoB (cont. var.) 1.0091 (1.0029,1.0153) 1.0078 (0.9878,1.0282)
##
## hyper: 1 vs 0 2.23 (1.63,3.05) 1.54 (1,2.38)
##
## db: 1 vs 0 1.89 (1.33,2.69) 1.16 (0.77,1.77)
##
## dyslip: 1 vs 0 1.53 (0.99,2.35) 0.8 (0.46,1.4)
##
## P(Wald's test) P(LR-test)
## sHCOT (cont. var.) 0.214 0.219
##
## age (cont. var.) 0.948 0.948
##
## sex: 2 vs 1 0.008 0.008
##
## ethnic: ref.=1 0.67
## 2 0.414
## 3 0.617
## 4 0.823
## 6 0.279
## 7 0.83
##
## educ: ref.=1 0.589
## 2 0.586
## 3 0.144
## 4 0.466
## 5 0.247
## 9 0.992
##
## masts: ref.=1 0.206
## 2 0.521
## 3 0.044
## 4 0.817
## 5 0.344
## 6 0.469
##
## finc (cont. var.) 0.097 0.094
##
## PIR (cont. var.) 0.629 0.629
##
## SMQ020: ref.=1 0.589
## 2 0.347
## 9 0.993
##
## ALQ101: ref.=1 0.493
## 2 0.371
## 9 0.985
##
## PAQ665: 2 vs 1 0.004 0.003
##
## PAQ650: 2 vs 1 0.134 0.12
##
## BMI (cont. var.) < 0.001 < 0.001
##
## SBP (cont. var.) 0.226 0.222
##
## DBP (cont. var.) 0.957 0.957
##
## AST (cont. var.) 0.81 0.812
##
## ALT (cont. var.) 0.532 0.539
##
## BUN (cont. var.) 0.565 0.565
##
## crea (cont. var.) 0.512 0.554
##
## TC (cont. var.) 0.572 0.572
##
## TG (cont. var.) 0.524 0.524
##
## HDL (cont. var.) 0.551 0.551
##
## LDL (cont. var.) 0.575 0.575
##
## apoB (cont. var.) 0.447 0.444
##
## hyper: 1 vs 0 0.052 0.052
##
## db: 1 vs 0 0.48 0.482
##
## dyslip: 1 vs 0 0.442 0.445
##
## Log-likelihood = -497.1138
## No. of observations = 1936
## AIC value = 1078.2276summary(Model5_3)##
## Call:
## glm(formula = dp10 ~ uCOT + age + sex + ethnic + educ + masts +
## finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 + BMI + SBP +
## DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB +
## hyper + db + dyslip, family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.6971 -0.3829 -0.2384 -0.1268 2.8235
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -3.730e+00 2.424e+00 -1.539 0.1238
## uCOT 1.403e-04 6.535e-05 2.147 0.0318 *
## age 1.536e-02 1.670e-02 0.920 0.3575
## sex2 4.950e-01 4.497e-01 1.101 0.2711
## ethnic2 6.795e-01 8.702e-01 0.781 0.4349
## ethnic3 1.317e+00 7.528e-01 1.749 0.0803 .
## ethnic4 1.196e+00 8.476e-01 1.411 0.1582
## ethnic6 1.325e+00 1.066e+00 1.242 0.2141
## ethnic7 -1.371e+01 1.161e+03 -0.012 0.9906
## educ2 -6.872e-01 7.661e-01 -0.897 0.3697
## educ3 -8.145e-01 7.115e-01 -1.145 0.2524
## educ4 -3.192e-01 6.875e-01 -0.464 0.6425
## educ5 -7.215e-01 8.111e-01 -0.890 0.3737
## masts2 2.236e-02 7.224e-01 0.031 0.9753
## masts3 7.478e-01 4.941e-01 1.514 0.1301
## masts4 4.881e-01 8.333e-01 0.586 0.5580
## masts5 -6.799e-01 6.100e-01 -1.115 0.2650
## masts6 -5.164e-01 7.778e-01 -0.664 0.5068
## finc -2.116e-01 1.258e-01 -1.682 0.0925 .
## PIR 2.911e-02 3.216e-01 0.091 0.9279
## SMQ0202 -3.432e-02 3.993e-01 -0.086 0.9315
## ALQ1012 -7.847e-01 4.678e-01 -1.678 0.0934 .
## PAQ6652 4.370e-01 3.968e-01 1.101 0.2707
## PAQ6502 3.834e-01 5.640e-01 0.680 0.4967
## BMI 4.201e-02 2.699e-02 1.557 0.1196
## SBP -3.325e-02 1.410e-02 -2.357 0.0184 *
## DBP 1.536e-02 1.601e-02 0.959 0.3376
## AST 2.781e-02 2.427e-02 1.146 0.2519
## ALT -1.426e-02 2.221e-02 -0.642 0.5209
## BUN 5.946e-03 3.586e-02 0.166 0.8683
## crea 5.883e-01 5.494e-01 1.071 0.2842
## TC -7.612e-01 6.288e-01 -1.210 0.2261
## TG 1.538e-01 1.258e-01 1.223 0.2213
## HDL 7.686e-01 6.288e-01 1.222 0.2216
## LDL 7.728e-01 6.290e-01 1.229 0.2192
## apoB -2.167e-03 2.187e-02 -0.099 0.9211
## hyper1 2.668e-01 4.690e-01 0.569 0.5695
## db1 1.675e-01 4.867e-01 0.344 0.7308
## dyslip1 -8.397e-01 5.861e-01 -1.433 0.1519
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 325.77 on 606 degrees of freedom
## Residual deviance: 254.44 on 568 degrees of freedom
## (4765 observations deleted due to missingness)
## AIC: 332.44
##
## Number of Fisher Scoring iterations: 16logistic.display(Model5_3)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI) P(Wald's test)
## uCOT (cont. var.) 1.0002 (1.0001,1.0003) 1.0001 (1,1.0003) 0.032
##
## age (cont. var.) 1.02 (1,1.03) 1.02 (0.98,1.05) 0.358
##
## sex: 2 vs 1 1.4 (0.76,2.57) 1.64 (0.68,3.96) 0.271
##
## ethnic: ref.=1
## 2 2.01 (0.43,9.33) 1.97 (0.36,10.86) 0.435
## 3 3.16 (0.94,10.66) 3.73 (0.85,16.31) 0.08
## 4 2.61 (0.68,9.93) 3.31 (0.63,17.41) 0.158
## 6 1.04 (0.17,6.44) 3.76 (0.47,30.41) 0.214
## 7 0 (0,Inf) 0 (0,Inf) 0.991
##
## educ: ref.=1
## 2 0.62 (0.18,2.16) 0.5 (0.11,2.26) 0.37
## 3 0.65 (0.21,2.01) 0.44 (0.11,1.79) 0.252
## 4 0.72 (0.25,2.05) 0.73 (0.19,2.8) 0.642
## 5 0.28 (0.08,0.98) 0.49 (0.1,2.38) 0.374
##
## masts: ref.=1
## 2 2.49 (0.78,7.87) 1.02 (0.25,4.21) 0.975
## 3 3.87 (1.76,8.51) 2.11 (0.8,5.56) 0.13
## 4 3.73 (1.14,12.17) 1.63 (0.32,8.34) 0.558
## 5 0.7 (0.26,1.94) 0.51 (0.15,1.67) 0.265
## 6 1.17 (0.33,4.14) 0.6 (0.13,2.74) 0.507
##
## finc (cont. var.) 0.83 (0.75,0.91) 0.81 (0.63,1.04) 0.093
##
## PIR (cont. var.) 0.65 (0.52,0.82) 1.03 (0.55,1.93) 0.928
##
## SMQ020: 2 vs 1 0.54 (0.29,1) 0.97 (0.44,2.11) 0.932
##
## ALQ101: 2 vs 1 0.89 (0.44,1.79) 0.46 (0.18,1.14) 0.093
##
## PAQ665: 2 vs 1 1.72 (0.9,3.3) 1.55 (0.71,3.37) 0.271
##
## PAQ650: 2 vs 1 2.55 (0.99,6.58) 1.47 (0.49,4.43) 0.497
##
## BMI (cont. var.) 1.03 (0.99,1.07) 1.04 (0.99,1.1) 0.12
##
## SBP (cont. var.) 0.99 (0.98,1.01) 0.97 (0.94,0.99) 0.018
##
## DBP (cont. var.) 1 (0.98,1.03) 1.02 (0.98,1.05) 0.338
##
## AST (cont. var.) 1.01 (0.99,1.04) 1.03 (0.98,1.08) 0.252
##
## ALT (cont. var.) 1 (0.98,1.02) 0.99 (0.94,1.03) 0.521
##
## BUN (cont. var.) 1.039 (0.9945,1.0854) 1.006 (0.9377,1.0792) 0.868
##
## crea (cont. var.) 1.92 (0.95,3.85) 1.8 (0.61,5.29) 0.284
##
## TC (cont. var.) 1 (1,1.01) 0.47 (0.14,1.6) 0.226
##
## TG (cont. var.) 1 (1,1.01) 1.17 (0.91,1.49) 0.221
##
## HDL (cont. var.) 0.99 (0.97,1.01) 2.16 (0.63,7.4) 0.222
##
## LDL (cont. var.) 1 (1,1.01) 2.17 (0.63,7.43) 0.219
##
## apoB (cont. var.) 1.0052 (0.9936,1.017) 0.9978 (0.956,1.0415) 0.921
##
## hyper: 1 vs 0 1.63 (0.89,2.98) 1.31 (0.52,3.27) 0.569
##
## db: 1 vs 0 1.29 (0.6,2.77) 1.18 (0.46,3.07) 0.731
##
## dyslip: 1 vs 0 1.06 (0.5,2.26) 0.43 (0.14,1.36) 0.152
##
## P(LR-test)
## uCOT (cont. var.) 0.026
##
## age (cont. var.) 0.359
##
## sex: 2 vs 1 0.27
##
## ethnic: ref.=1 0.379
## 2
## 3
## 4
## 6
## 7
##
## educ: ref.=1 0.705
## 2
## 3
## 4
## 5
##
## masts: ref.=1 0.343
## 2
## 3
## 4
## 5
## 6
##
## finc (cont. var.) 0.081
##
## PIR (cont. var.) 0.928
##
## SMQ020: 2 vs 1 0.932
##
## ALQ101: 2 vs 1 0.083
##
## PAQ665: 2 vs 1 0.265
##
## PAQ650: 2 vs 1 0.486
##
## BMI (cont. var.) 0.124
##
## SBP (cont. var.) 0.014
##
## DBP (cont. var.) 0.321
##
## AST (cont. var.) 0.267
##
## ALT (cont. var.) 0.51
##
## BUN (cont. var.) 0.869
##
## crea (cont. var.) 0.29
##
## TC (cont. var.) 0.222
##
## TG (cont. var.) 0.217
##
## HDL (cont. var.) 0.217
##
## LDL (cont. var.) 0.215
##
## apoB (cont. var.) 0.921
##
## hyper: 1 vs 0 0.569
##
## db: 1 vs 0 0.732
##
## dyslip: 1 vs 0 0.156
##
## Log-likelihood = -127.2205
## No. of observations = 607
## AIC value = 332.441summary(Model5_4)##
## Call:
## glm(formula = dp10 ~ uHCOT + age + sex + ethnic + educ + masts +
## finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 + BMI + SBP +
## DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB +
## hyper + db + dyslip, family = binomial, data = tab1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.5166 -0.3912 -0.2371 -0.1291 2.7453
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -3.258e+00 2.397e+00 -1.359 0.1741
## uHCOT 3.172e-05 2.217e-05 1.431 0.1525
## age 1.088e-02 1.640e-02 0.664 0.5070
## sex2 3.990e-01 4.452e-01 0.896 0.3702
## ethnic2 7.091e-01 8.659e-01 0.819 0.4128
## ethnic3 1.409e+00 7.523e-01 1.873 0.0611 .
## ethnic4 1.264e+00 8.461e-01 1.494 0.1353
## ethnic6 1.309e+00 1.065e+00 1.229 0.2192
## ethnic7 -1.263e+01 7.070e+02 -0.018 0.9857
## educ2 -6.642e-01 7.668e-01 -0.866 0.3864
## educ3 -7.085e-01 7.016e-01 -1.010 0.3125
## educ4 -3.121e-01 6.880e-01 -0.454 0.6501
## educ5 -7.199e-01 8.124e-01 -0.886 0.3756
## masts2 3.671e-02 7.233e-01 0.051 0.9595
## masts3 7.959e-01 4.898e-01 1.625 0.1042
## masts4 5.879e-01 8.263e-01 0.712 0.4768
## masts5 -6.857e-01 6.086e-01 -1.127 0.2599
## masts6 -4.917e-01 7.761e-01 -0.634 0.5264
## finc -2.009e-01 1.251e-01 -1.605 0.1084
## PIR 1.354e-03 3.200e-01 0.004 0.9966
## SMQ0202 -1.173e-01 3.929e-01 -0.299 0.7652
## ALQ1012 -6.793e-01 4.562e-01 -1.489 0.1364
## PAQ6652 4.156e-01 3.940e-01 1.055 0.2915
## PAQ6502 4.501e-01 5.625e-01 0.800 0.4236
## BMI 3.818e-02 2.687e-02 1.421 0.1554
## SBP -3.420e-02 1.414e-02 -2.419 0.0156 *
## DBP 1.616e-02 1.618e-02 0.999 0.3177
## AST 2.786e-02 2.414e-02 1.154 0.2484
## ALT -1.514e-02 2.230e-02 -0.679 0.4971
## BUN 9.932e-03 3.560e-02 0.279 0.7802
## crea 4.984e-01 5.441e-01 0.916 0.3597
## TC -8.987e-01 6.213e-01 -1.446 0.1480
## TG 1.809e-01 1.243e-01 1.456 0.1454
## HDL 9.042e-01 6.214e-01 1.455 0.1456
## LDL 9.113e-01 6.214e-01 1.466 0.1425
## apoB -2.895e-03 2.166e-02 -0.134 0.8937
## hyper1 3.358e-01 4.642e-01 0.723 0.4695
## db1 1.546e-01 4.868e-01 0.318 0.7508
## dyslip1 -7.751e-01 5.793e-01 -1.338 0.1809
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 325.77 on 606 degrees of freedom
## Residual deviance: 257.51 on 568 degrees of freedom
## (4765 observations deleted due to missingness)
## AIC: 335.51
##
## Number of Fisher Scoring iterations: 15logistic.display(Model5_4)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## uHCOT (cont. var.) 1 (1,1.0001) 1 (1,1.0001)
##
## age (cont. var.) 1.02 (1,1.03) 1.01 (0.98,1.04)
##
## sex: 2 vs 1 1.4 (0.76,2.57) 1.49 (0.62,3.57)
##
## ethnic: ref.=1
## 2 2.01 (0.43,9.33) 2.03 (0.37,11.09)
## 3 3.16 (0.94,10.66) 4.09 (0.94,17.87)
## 4 2.61 (0.68,9.93) 3.54 (0.67,18.58)
## 6 1.04 (0.17,6.44) 3.7 (0.46,29.88)
## 7 0 (0,Inf) 0 (0,Inf)
##
## educ: ref.=1
## 2 0.62 (0.18,2.16) 0.51 (0.11,2.31)
## 3 0.65 (0.21,2.01) 0.49 (0.12,1.95)
## 4 0.72 (0.25,2.05) 0.73 (0.19,2.82)
## 5 0.28 (0.08,0.98) 0.49 (0.1,2.39)
##
## masts: ref.=1
## 2 2.49 (0.78,7.87) 1.04 (0.25,4.28)
## 3 3.87 (1.76,8.51) 2.22 (0.85,5.79)
## 4 3.73 (1.14,12.17) 1.8 (0.36,9.09)
## 5 0.7 (0.26,1.94) 0.5 (0.15,1.66)
## 6 1.17 (0.33,4.14) 0.61 (0.13,2.8)
##
## finc (cont. var.) 0.83 (0.75,0.91) 0.82 (0.64,1.05)
##
## PIR (cont. var.) 0.6519 (0.518,0.8204) 1.0014 (0.5348,1.8751)
##
## SMQ020: 2 vs 1 0.54 (0.29,1) 0.89 (0.41,1.92)
##
## ALQ101: 2 vs 1 0.89 (0.44,1.79) 0.51 (0.21,1.24)
##
## PAQ665: 2 vs 1 1.72 (0.9,3.3) 1.52 (0.7,3.28)
##
## PAQ650: 2 vs 1 2.55 (0.99,6.58) 1.57 (0.52,4.72)
##
## BMI (cont. var.) 1.03 (0.99,1.07) 1.04 (0.99,1.1)
##
## SBP (cont. var.) 0.99 (0.98,1.01) 0.97 (0.94,0.99)
##
## DBP (cont. var.) 1 (0.98,1.03) 1.02 (0.98,1.05)
##
## AST (cont. var.) 1.01 (0.99,1.04) 1.03 (0.98,1.08)
##
## ALT (cont. var.) 1 (0.98,1.02) 0.98 (0.94,1.03)
##
## BUN (cont. var.) 1.039 (0.9945,1.0854) 1.01 (0.9419,1.083)
##
## crea (cont. var.) 1.92 (0.95,3.85) 1.65 (0.57,4.78)
##
## TC (cont. var.) 1 (1,1.01) 0.41 (0.12,1.38)
##
## TG (cont. var.) 1 (1,1.01) 1.2 (0.94,1.53)
##
## HDL (cont. var.) 0.99 (0.97,1.01) 2.47 (0.73,8.35)
##
## LDL (cont. var.) 1 (1,1.01) 2.49 (0.74,8.41)
##
## apoB (cont. var.) 1.0052 (0.9936,1.017) 0.9971 (0.9557,1.0404)
##
## hyper: 1 vs 0 1.63 (0.89,2.98) 1.4 (0.56,3.48)
##
## db: 1 vs 0 1.29 (0.6,2.77) 1.17 (0.45,3.03)
##
## dyslip: 1 vs 0 1.06 (0.5,2.26) 0.46 (0.15,1.43)
##
## P(Wald's test) P(LR-test)
## uHCOT (cont. var.) 0.153 0.169
##
## age (cont. var.) 0.507 0.508
##
## sex: 2 vs 1 0.37 0.369
##
## ethnic: ref.=1 0.319
## 2 0.413
## 3 0.061
## 4 0.135
## 6 0.219
## 7 0.986
##
## educ: ref.=1 0.777
## 2 0.386
## 3 0.313
## 4 0.65
## 5 0.376
##
## masts: ref.=1 0.29
## 2 0.96
## 3 0.104
## 4 0.477
## 5 0.26
## 6 0.526
##
## finc (cont. var.) 0.108 0.097
##
## PIR (cont. var.) 0.997 0.997
##
## SMQ020: 2 vs 1 0.765 0.765
##
## ALQ101: 2 vs 1 0.136 0.126
##
## PAQ665: 2 vs 1 0.292 0.286
##
## PAQ650: 2 vs 1 0.424 0.41
##
## BMI (cont. var.) 0.155 0.159
##
## SBP (cont. var.) 0.016 0.012
##
## DBP (cont. var.) 0.318 0.3
##
## AST (cont. var.) 0.248 0.263
##
## ALT (cont. var.) 0.497 0.486
##
## BUN (cont. var.) 0.78 0.781
##
## crea (cont. var.) 0.36 0.365
##
## TC (cont. var.) 0.148 0.143
##
## TG (cont. var.) 0.145 0.141
##
## HDL (cont. var.) 0.146 0.141
##
## LDL (cont. var.) 0.143 0.138
##
## apoB (cont. var.) 0.894 0.894
##
## hyper: 1 vs 0 0.47 0.468
##
## db: 1 vs 0 0.751 0.752
##
## dyslip: 1 vs 0 0.181 0.186
##
## Log-likelihood = -128.755
## No. of observations = 607
## AIC value = 335.519. Filter out values ‘7, 77’ and ‘9, 99’ from categorical variables column
filter_edu <- tab1 %>% filter(educ %in% c('1', '2', '3', '4', '5'))
filter_masts <- filter_edu %>% filter(masts %in% c('1', '2', '3', '4', '5', '6'))
filter_alc <- filter_masts %>% filter(ALQ101 %in% c('1', '2'))
filter_smk <- filter_alc %>% filter(SMQ020 %in% c('1', '2'))# Descriptive analyis
comp = compareGroups(dp10 ~ age + sex + ethnic + educ + masts + SBP + DBP + BMI + finc + PIR +
BUN + crea + AST + ALT + apoB + HDL + LDL + TG + TC +
Glu + tolGlu + ALQ101 + SMQ020 + PAQ650 + PAQ665 +
sCOT + sHCOT + uCOT + uHCOT + db + dyslip + hyper , data = filter_smk)## Warning in compare.i(X[, i], y = y, selec.i = selec[i], method.i = method[i], :
## Some levels of 'educ' are removed since no observation in that/those levels## Warning in compare.i(X[, i], y = y, selec.i = selec[i], method.i = method[i], :
## Some levels of 'masts' are removed since no observation in that/those levels## Warning in compare.i(X[, i], y = y, selec.i = selec[i], method.i = method[i], :
## Some levels of 'ALQ101' are removed since no observation in that/those levels## Warning in compare.i(X[, i], y = y, selec.i = selec[i], method.i = method[i], :
## Some levels of 'SMQ020' are removed since no observation in that/those levelsCheck normal distribution with ggplot2
# Depressive disorder distribution
ggplot(data = filter_smk, mapping = aes(dp10, fill = dp10)) + geom_bar(width = 0.5) +
xlab("Depressive Disorder") +
scale_x_discrete(labels = c ("No", "Yes")) +
ylab("Count") +
theme(plot.title = element_text(hjust = 0.5)) +
theme(legend.position = "none") +
geom_text(stat='count', aes(label=..count..), vjust=-0.5) +
geom_text(stat='count', aes(label=paste0(round(after_stat(count/sum(count)*100),1),'%')), vjust=2, color = "black") ## Warning: The dot-dot notation (`..count..`) was deprecated in ggplot2 3.4.0.
## i Please use `after_stat(count)` instead.
# theme_economist() fivenum(filter_smk$sCOT)## [1] 0.011 0.011 0.036 10.900 1820.000fivenum(filter_smk$sHCOT)## [1] 0.011 0.011 0.011 3.680 1150.000fivenum(filter_smk$uCOT)## [1] 2.10e-02 1.43e-01 4.51e-01 1.31e+02 2.68e+04fivenum(filter_smk$uHCOT)## [1] 2.10e-02 2.31e-01 8.29e-01 2.08e+02 7.06e+04# Independent variables
ggplot(data = filter_smk, mapping = aes(sCOT)) +
geom_histogram(fill = "blue", color = "white")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 175 rows containing non-finite values (`stat_bin()`).
ggplot(data = filter_smk, mapping = aes(sHCOT)) +
geom_histogram(fill = "blue", color = "white")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 175 rows containing non-finite values (`stat_bin()`).
ggplot(data = filter_smk, mapping = aes(uCOT)) +
geom_histogram(fill = "blue", color = "white")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 3430 rows containing non-finite values (`stat_bin()`).
ggplot(data = filter_smk, mapping = aes(uHCOT)) +
geom_histogram(fill = "blue", color = "white")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 3430 rows containing non-finite values (`stat_bin()`).
# Covariates
ggplot(data = filter_smk, mapping = aes(SBP)) +
geom_histogram(fill = "blue", color = "white")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 467 rows containing non-finite values (`stat_bin()`).
ggplot(data = filter_smk, mapping = aes(DBP)) +
geom_histogram(fill = "blue", color = "white")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 467 rows containing non-finite values (`stat_bin()`).
ggplot(data = filter_smk, mapping = aes(finc)) +
geom_histogram(fill = "blue", color = "white")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 49 rows containing non-finite values (`stat_bin()`).
ggplot(data = filter_smk, mapping = aes(PIR)) +
geom_histogram(fill = "blue", color = "white")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 364 rows containing non-finite values (`stat_bin()`).
ggplot(data = filter_smk, mapping = aes(BUN)) +
geom_histogram(fill = "blue", color = "white")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 190 rows containing non-finite values (`stat_bin()`).
ggplot(data = filter_smk, mapping = aes(crea)) +
geom_histogram(fill = "blue", color = "white")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 190 rows containing non-finite values (`stat_bin()`).
ggplot(data = filter_smk, mapping = aes(AST)) +
geom_histogram(fill = "blue", color = "white")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 192 rows containing non-finite values (`stat_bin()`).
ggplot(data = filter_smk, mapping = aes(ALT)) +
geom_histogram(fill = "blue", color = "white")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 192 rows containing non-finite values (`stat_bin()`).
ggplot(data = filter_smk, mapping = aes(TC)) +
geom_histogram(fill = "blue", color = "white")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 176 rows containing non-finite values (`stat_bin()`).
ggplot(data = filter_smk, mapping = aes(TG)) +
geom_histogram(fill = "blue", color = "white")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 2719 rows containing non-finite values (`stat_bin()`).
ggplot(data = filter_smk, mapping = aes(HDL)) +
geom_histogram(fill = "blue", color = "white")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 176 rows containing non-finite values (`stat_bin()`).
ggplot(data = filter_smk, mapping = aes(LDL)) +
geom_histogram(fill = "blue", color = "white")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 2756 rows containing non-finite values (`stat_bin()`).
ggplot(data = filter_smk, mapping = aes(apoB)) +
geom_histogram(fill = "blue", color = "white")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 2720 rows containing non-finite values (`stat_bin()`).
ggplot(data = filter_smk, mapping = aes(Glu)) +
geom_histogram(fill = "blue", color = "white")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 2705 rows containing non-finite values (`stat_bin()`).
createTable(comp)##
## --------Summary descriptives table by 'dp10'---------
##
## ________________________________________________________________________
## 0 1 p.overall
## N=4553 N=487
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
## Age in years at screening 48.9 (17.6) 51.6 (16.4) 0.001
## sex: <0.001
## 1 2275 (50.0%) 158 (32.4%)
## 2 2278 (50.0%) 329 (67.6%)
## ethnic: <0.001
## 1 607 (13.3%) 67 (13.8%)
## 2 392 (8.61%) 57 (11.7%)
## 3 2006 (44.1%) 216 (44.4%)
## 4 905 (19.9%) 105 (21.6%)
## 6 509 (11.2%) 17 (3.49%)
## 7 134 (2.94%) 25 (5.13%)
## educ: <0.001
## 1 292 (6.41%) 64 (13.1%)
## 2 575 (12.6%) 103 (21.1%)
## 3 1009 (22.2%) 117 (24.0%)
## 4 1442 (31.7%) 148 (30.4%)
## 5 1235 (27.1%) 55 (11.3%)
## masts: <0.001
## 1 2443 (53.7%) 187 (38.4%)
## 2 309 (6.79%) 57 (11.7%)
## 3 484 (10.6%) 98 (20.1%)
## 4 123 (2.70%) 27 (5.54%)
## 5 872 (19.2%) 82 (16.8%)
## 6 322 (7.07%) 36 (7.39%)
## Systolic: Blood pres (1st rdg) mm Hg 123 (17.2) 125 (19.1) 0.003
## Diastolic: Blood pres (1st rdg) mm Hg 69.4 (12.5) 69.7 (12.4) 0.656
## Body Mass Index (kg/m**2) 28.9 (6.87) 31.9 (9.32) <0.001
## Annual family income 10.8 (13.4) 8.97 (15.0) 0.010
## Ratio of family income to poverty 2.62 (1.66) 1.74 (1.33) <0.001
## Blood urea nitrogen (mg/dL) 13.4 (5.73) 13.6 (9.17) 0.597
## Creatinine (mg/dL) 0.91 (0.44) 0.99 (1.09) 0.103
## Aspartate aminotransferase AST (U/L) 25.3 (18.6) 26.6 (21.7) 0.216
## Alanine aminotransferase ALT (U/L) 25.0 (18.9) 25.7 (21.5) 0.521
## Apolipoprotein (B) (mg/dL) 89.8 (24.6) 95.3 (26.5) 0.004
## Direct HDL-Cholesterol (mg/dL) 52.9 (16.2) 51.2 (15.5) 0.019
## LDL-cholesterol (mg/dL) 111 (34.6) 113 (37.1) 0.359
## Triglyceride (mg/dL) 120 (130) 142 (98.7) 0.002
## Total Cholesterol( mg/dL) 189 (41.9) 192 (41.3) 0.090
## Fasting Glucose (mg/dL) 107 (32.6) 117 (43.8) 0.001
## Two Hour Glucose(OGTT) (mg/dL) 118 (50.1) 129 (52.3) 0.016
## ALQ101: 0.872
## 1 3293 (72.3%) 350 (71.9%)
## 2 1260 (27.7%) 137 (28.1%)
## SMQ020: <0.001
## 1 1924 (42.3%) 280 (57.5%)
## 2 2629 (57.7%) 207 (42.5%)
## PAQ650: <0.001
## 1 1090 (23.9%) 48 (9.86%)
## 2 3463 (76.1%) 439 (90.1%)
## PAQ665: <0.001
## 1 2009 (44.1%) 123 (25.3%)
## 2 2544 (55.9%) 364 (74.7%)
## Cotinine, Serum (ng/mL) 56.7 (131) 104 (163) <0.001
## Hematocrit (%) 22.6 (60.9) 45.4 (78.0) <0.001
## Total Cotinine, urine (ng/mL) 686 (1833) 1573 (3242) 0.002
## Total Hydroxycotinine, urine (ng/mL) 1348 (4209) 3143 (6196) 0.001
## db: <0.001
## 0 3849 (84.5%) 364 (74.7%)
## 1 704 (15.5%) 123 (25.3%)
## dyslip: 0.013
## 0 1354 (29.7%) 118 (24.2%)
## 1 3199 (70.3%) 369 (75.8%)
## hyper: <0.001
## 0 2915 (64.0%) 225 (46.2%)
## 1 1638 (36.0%) 262 (53.8%)
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯# Describe by median and IQR, for example: age
med_iqr <- compareGroups(dp10 ~ . -ALQ110, data = filter_smk, method = c(age = 2, finc = 2, PIR = 2, crea = 2, AST = 2, ALT = 2, TG = 2, Glu = 2))## Warning in compare.i(X[, i], y = y, selec.i = selec[i], method.i = method[i], :
## Some levels of 'educ' are removed since no observation in that/those levels## Warning in compare.i(X[, i], y = y, selec.i = selec[i], method.i = method[i], :
## Some levels of 'masts' are removed since no observation in that/those levels## Warning in compare.i(X[, i], y = y, selec.i = selec[i], method.i = method[i], :
## Some levels of 'ALQ101' are removed since no observation in that/those levels## Warning in compare.i(X[, i], y = y, selec.i = selec[i], method.i = method[i], :
## Some levels of 'SMQ020' are removed since no observation in that/those levelsmed_iqr##
##
## -------- Summary of results by groups of 'dp10'---------
##
##
## var N p.value method
## 1 Age in years at screening 5040 0.001** continuous non-normal
## 2 sex 5040 <0.001** categorical
## 3 ethnic 5040 <0.001** categorical
## 4 educ 5040 <0.001** categorical
## 5 masts 5040 <0.001** categorical
## 6 Systolic: Blood pres (1st rdg) mm Hg 4573 0.003** continuous normal
## 7 Diastolic: Blood pres (1st rdg) mm Hg 4573 0.656 continuous normal
## 8 Body Mass Index (kg/m**2) 4994 <0.001** continuous normal
## 9 Annual family income 4991 <0.001** continuous non-normal
## 10 Ratio of family income to poverty 4676 <0.001** continuous non-normal
## 11 Blood urea nitrogen (mg/dL) 4850 0.597 continuous normal
## 12 Creatinine (mg/dL) 4850 0.052* continuous non-normal
## 13 Aspartate aminotransferase AST (U/L) 4848 0.467 continuous non-normal
## 14 Alanine aminotransferase ALT (U/L) 4848 0.369 continuous non-normal
## 15 Apolipoprotein (B) (mg/dL) 2320 0.004** continuous normal
## 16 Direct HDL-Cholesterol (mg/dL) 4864 0.019** continuous normal
## 17 LDL-cholesterol (mg/dL) 2284 0.359 continuous normal
## 18 Triglyceride (mg/dL) 2321 <0.001** continuous non-normal
## 19 Total Cholesterol( mg/dL) 4864 0.090* continuous normal
## 20 Fasting Glucose (mg/dL) 2335 <0.001** continuous non-normal
## 21 Two Hour Glucose(OGTT) (mg/dL) 1804 0.016** continuous normal
## 22 ALQ101 5040 0.872 categorical
## 23 SMQ020 5040 <0.001** categorical
## 24 PAQ665 5040 <0.001** categorical
## 25 PAQ650 5040 <0.001** categorical
## 26 Cotinine, Serum (ng/mL) 4865 <0.001** continuous normal
## 27 Hematocrit (%) 4865 <0.001** continuous normal
## 28 Total Cotinine, urine (ng/mL) 1610 0.002** continuous normal
## 29 Total Hydroxycotinine, urine (ng/mL) 1610 0.001** continuous normal
## 30 db 5040 <0.001** categorical
## 31 dyslip 5040 0.013** categorical
## 32 hyper 5040 <0.001** categorical
## selection
## 1 ALL
## 2 ALL
## 3 ALL
## 4 ALL
## 5 ALL
## 6 ALL
## 7 ALL
## 8 ALL
## 9 ALL
## 10 ALL
## 11 ALL
## 12 ALL
## 13 ALL
## 14 ALL
## 15 ALL
## 16 ALL
## 17 ALL
## 18 ALL
## 19 ALL
## 20 ALL
## 21 ALL
## 22 ALL
## 23 ALL
## 24 ALL
## 25 ALL
## 26 ALL
## 27 ALL
## 28 ALL
## 29 ALL
## 30 ALL
## 31 ALL
## 32 ALL
## -----
## Signif. codes: 0 '**' 0.05 '*' 0.1 ' ' 1createTable(med_iqr)##
## --------Summary descriptives table by 'dp10'---------
##
## _________________________________________________________________________________
## 0 1 p.overall
## N=4553 N=487
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
## Age in years at screening 48.0 [34.0;63.0] 54.0 [38.0;63.0] 0.001
## sex: <0.001
## 1 2275 (50.0%) 158 (32.4%)
## 2 2278 (50.0%) 329 (67.6%)
## ethnic: <0.001
## 1 607 (13.3%) 67 (13.8%)
## 2 392 (8.61%) 57 (11.7%)
## 3 2006 (44.1%) 216 (44.4%)
## 4 905 (19.9%) 105 (21.6%)
## 6 509 (11.2%) 17 (3.49%)
## 7 134 (2.94%) 25 (5.13%)
## educ: <0.001
## 1 292 (6.41%) 64 (13.1%)
## 2 575 (12.6%) 103 (21.1%)
## 3 1009 (22.2%) 117 (24.0%)
## 4 1442 (31.7%) 148 (30.4%)
## 5 1235 (27.1%) 55 (11.3%)
## masts: <0.001
## 1 2443 (53.7%) 187 (38.4%)
## 2 309 (6.79%) 57 (11.7%)
## 3 484 (10.6%) 98 (20.1%)
## 4 123 (2.70%) 27 (5.54%)
## 5 872 (19.2%) 82 (16.8%)
## 6 322 (7.07%) 36 (7.39%)
## Systolic: Blood pres (1st rdg) mm Hg 123 (17.2) 125 (19.1) 0.003
## Diastolic: Blood pres (1st rdg) mm Hg 69.4 (12.5) 69.7 (12.4) 0.656
## Body Mass Index (kg/m**2) 28.9 (6.87) 31.9 (9.32) <0.001
## Annual family income 8.00 [5.00;14.0] 6.00 [3.00;8.00] <0.001
## Ratio of family income to poverty 2.26 [1.13;4.28] 1.27 [0.78;2.35] <0.001
## Blood urea nitrogen (mg/dL) 13.4 (5.73) 13.6 (9.17) 0.597
## Creatinine (mg/dL) 0.86 [0.72;1.02] 0.82 [0.71;1.01] 0.052
## Aspartate aminotransferase AST (U/L) 22.0 [19.0;27.0] 22.0 [18.5;28.0] 0.467
## Alanine aminotransferase ALT (U/L) 20.0 [16.0;28.0] 20.0 [15.0;28.0] 0.369
## Apolipoprotein (B) (mg/dL) 89.8 (24.6) 95.3 (26.5) 0.004
## Direct HDL-Cholesterol (mg/dL) 52.9 (16.2) 51.2 (15.5) 0.019
## LDL-cholesterol (mg/dL) 111 (34.6) 113 (37.1) 0.359
## Triglyceride (mg/dL) 94.0 [64.0;139] 120 [80.0;183] <0.001
## Total Cholesterol( mg/dL) 189 (41.9) 192 (41.3) 0.090
## Fasting Glucose (mg/dL) 99.0 [92.0;108] 102 [95.0;117] <0.001
## Two Hour Glucose(OGTT) (mg/dL) 118 (50.1) 129 (52.3) 0.016
## ALQ101: 0.872
## 1 3293 (72.3%) 350 (71.9%)
## 2 1260 (27.7%) 137 (28.1%)
## SMQ020: <0.001
## 1 1924 (42.3%) 280 (57.5%)
## 2 2629 (57.7%) 207 (42.5%)
## PAQ665: <0.001
## 1 2009 (44.1%) 123 (25.3%)
## 2 2544 (55.9%) 364 (74.7%)
## PAQ650: <0.001
## 1 1090 (23.9%) 48 (9.86%)
## 2 3463 (76.1%) 439 (90.1%)
## Cotinine, Serum (ng/mL) 56.7 (131) 104 (163) <0.001
## Hematocrit (%) 22.6 (60.9) 45.4 (78.0) <0.001
## Total Cotinine, urine (ng/mL) 686 (1833) 1573 (3242) 0.002
## Total Hydroxycotinine, urine (ng/mL) 1348 (4209) 3143 (6196) 0.001
## db: <0.001
## 0 3849 (84.5%) 364 (74.7%)
## 1 704 (15.5%) 123 (25.3%)
## dyslip: 0.013
## 0 1354 (29.7%) 118 (24.2%)
## 1 3199 (70.3%) 369 (75.8%)
## hyper: <0.001
## 0 2915 (64.0%) 225 (46.2%)
## 1 1638 (36.0%) 262 (53.8%)
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯ggplot(filter_smk, aes(x = dp10, y = sCOT, fill = dp10)) + geom_boxplot() +
xlab("Depressive Disorder") +
scale_x_discrete(labels = c ("No", "Yes")) +
ylab("Serum Cotinine") +
ggtitle("Relationship between Serum Cotinine & Depressive Disorder") +
theme(plot.title = element_text(hjust = 0.5)) +
theme(legend.position = "none")## Warning: Removed 175 rows containing non-finite values (`stat_boxplot()`).
# Check logistic regression
dat = filter_smk
mo_sCOT = glm(dp10 ~ sCOT + age + sex + ethnic + educ + masts + finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 + BMI + SBP + DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB + hyper + db + dyslip, family = binomial, data = dat)
mo_sHCOT = glm(dp10 ~ sHCOT + age + sex + ethnic + educ + masts + finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 +BMI + SBP + DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB + hyper + db + dyslip, family = binomial, data = dat)
mo_uCOT = glm(dp10 ~ uCOT + age + sex + ethnic + educ + masts + finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 + BMI + SBP + DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB + hyper + db + dyslip, family = binomial, data = dat)
mo_sHCOT = glm(dp10 ~ uHCOT + age + sex + ethnic + educ + masts + finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 + BMI + SBP + DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB + hyper + db + dyslip, family = binomial, data = dat)
# Summary the results of logistic regression
summary(mo_sCOT)##
## Call:
## glm(formula = dp10 ~ sCOT + age + sex + ethnic + educ + masts +
## finc + PIR + SMQ020 + ALQ101 + PAQ665 + PAQ650 + BMI + SBP +
## DBP + AST + ALT + BUN + crea + TC + TG + HDL + LDL + apoB +
## hyper + db + dyslip, family = binomial, data = dat)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.4731 -0.4650 -0.3023 -0.1811 3.4276
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -4.700e+00 1.103e+00 -4.260 2.04e-05 ***
## sCOT 1.010e-03 5.661e-04 1.784 0.074491 .
## age -8.583e-05 8.146e-03 -0.011 0.991594
## sex2 5.667e-01 2.107e-01 2.690 0.007150 **
## ethnic2 2.774e-01 3.536e-01 0.785 0.432708
## ethnic3 1.196e-01 2.893e-01 0.413 0.679301
## ethnic4 2.893e-02 3.368e-01 0.086 0.931538
## ethnic6 -6.089e-01 5.338e-01 -1.141 0.253996
## ethnic7 9.657e-02 6.241e-01 0.155 0.877034
## educ2 -2.001e-01 3.487e-01 -0.574 0.565943
## educ3 -5.077e-01 3.511e-01 -1.446 0.148106
## educ4 -2.371e-01 3.384e-01 -0.701 0.483454
## educ5 -4.483e-01 4.089e-01 -1.096 0.272917
## masts2 2.103e-01 3.219e-01 0.653 0.513450
## masts3 5.003e-01 2.495e-01 2.005 0.044914 *
## masts4 9.218e-02 4.385e-01 0.210 0.833507
## masts5 -3.002e-01 2.968e-01 -1.012 0.311651
## masts6 -2.638e-01 3.641e-01 -0.725 0.468706
## finc -9.393e-02 5.738e-02 -1.637 0.101657
## PIR -7.574e-02 1.503e-01 -0.504 0.614274
## SMQ0202 -1.345e-01 1.960e-01 -0.686 0.492474
## ALQ1012 -1.881e-01 2.130e-01 -0.883 0.377299
## PAQ6652 5.698e-01 1.976e-01 2.884 0.003930 **
## PAQ6502 4.366e-01 2.960e-01 1.475 0.140151
## BMI 3.998e-02 1.171e-02 3.415 0.000637 ***
## SBP -7.295e-03 5.875e-03 -1.242 0.214317
## DBP 1.651e-04 7.360e-03 0.022 0.982099
## AST 9.757e-04 3.618e-03 0.270 0.787439
## ALT 3.905e-03 6.221e-03 0.628 0.530166
## BUN 1.105e-02 1.657e-02 0.667 0.504631
## crea 1.159e-01 1.812e-01 0.640 0.522391
## TC -1.756e-01 2.998e-01 -0.586 0.558126
## TG 3.956e-02 5.992e-02 0.660 0.509095
## HDL 1.861e-01 2.998e-01 0.621 0.534722
## LDL 1.740e-01 2.995e-01 0.581 0.561383
## apoB 7.840e-03 1.025e-02 0.765 0.444177
## hyper1 4.334e-01 2.223e-01 1.950 0.051224 .
## db1 1.497e-01 2.134e-01 0.702 0.482940
## dyslip1 -2.129e-01 2.827e-01 -0.753 0.451287
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1169.18 on 1929 degrees of freedom
## Residual deviance: 992.63 on 1891 degrees of freedom
## (3110 observations deleted due to missingness)
## AIC: 1070.6
##
## Number of Fisher Scoring iterations: 6logistic.display(mo_sCOT)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## sCOT (cont. var.) 1.0015 (1.0006,1.0024) 1.001 (0.9999,1.0021)
##
## age (cont. var.) 1.0152 (1.0062,1.0244) 0.9999 (0.9841,1.016)
##
## sex: 2 vs 1 1.74 (1.26,2.41) 1.76 (1.17,2.66)
##
## ethnic: ref.=1
## 2 1.16 (0.61,2.19) 1.32 (0.66,2.64)
## 3 1 (0.62,1.61) 1.13 (0.64,1.99)
## 4 0.89 (0.51,1.56) 1.03 (0.53,1.99)
## 6 0.22 (0.08,0.59) 0.54 (0.19,1.55)
## 7 0.89 (0.29,2.69) 1.1 (0.32,3.74)
##
## educ: ref.=1
## 2 0.66 (0.36,1.19) 0.82 (0.41,1.62)
## 3 0.45 (0.25,0.8) 0.6 (0.3,1.2)
## 4 0.52 (0.3,0.89) 0.79 (0.41,1.53)
## 5 0.2 (0.11,0.38) 0.64 (0.29,1.42)
##
## masts: ref.=1
## 2 2.34 (1.37,3.97) 1.23 (0.66,2.32)
## 3 2.73 (1.78,4.19) 1.65 (1.01,2.69)
## 4 2.43 (1.15,5.14) 1.1 (0.46,2.59)
## 5 0.87 (0.53,1.42) 0.74 (0.41,1.32)
## 6 1.23 (0.65,2.33) 0.77 (0.38,1.57)
##
## finc (cont. var.) 0.86 (0.82,0.89) 0.91 (0.81,1.02)
##
## PIR (cont. var.) 0.69 (0.61,0.77) 0.93 (0.69,1.24)
##
## SMQ020: 2 vs 1 0.59 (0.43,0.81) 0.87 (0.6,1.28)
##
## ALQ101: 2 vs 1 1.03 (0.72,1.46) 0.83 (0.55,1.26)
##
## PAQ665: 2 vs 1 2.44 (1.71,3.49) 1.77 (1.2,2.6)
##
## PAQ650: 2 vs 1 3.25 (1.92,5.5) 1.55 (0.87,2.76)
##
## BMI (cont. var.) 1.06 (1.04,1.08) 1.04 (1.02,1.06)
##
## SBP (cont. var.) 1.0088 (1.0004,1.0172) 0.9927 (0.9814,1.0042)
##
## DBP (cont. var.) 1.001 (0.9886,1.0135) 1.0002 (0.9858,1.0147)
##
## AST (cont. var.) 1.0026 (0.9985,1.0066) 1.001 (0.9939,1.0081)
##
## ALT (cont. var.) 1.005 (0.998,1.0121) 1.0039 (0.9917,1.0162)
##
## BUN (cont. var.) 1.03 (1,1.05) 1.01 (0.98,1.04)
##
## crea (cont. var.) 1.07 (0.85,1.35) 1.12 (0.79,1.6)
##
## TC (cont. var.) 1 (1,1.01) 0.84 (0.47,1.51)
##
## TG (cont. var.) 1.01 (1,1.01) 1.04 (0.93,1.17)
##
## HDL (cont. var.) 0.99 (0.98,1) 1.2 (0.67,2.17)
##
## LDL (cont. var.) 1 (1,1.01) 1.19 (0.66,2.14)
##
## apoB (cont. var.) 1.009 (1.0029,1.0152) 1.0079 (0.9878,1.0283)
##
## hyper: 1 vs 0 2.24 (1.63,3.07) 1.54 (1,2.38)
##
## db: 1 vs 0 1.89 (1.33,2.69) 1.16 (0.76,1.76)
##
## dyslip: 1 vs 0 1.54 (1,2.37) 0.81 (0.46,1.41)
##
## P(Wald's test) P(LR-test)
## sCOT (cont. var.) 0.074 0.078
##
## age (cont. var.) 0.992 0.992
##
## sex: 2 vs 1 0.007 0.007
##
## ethnic: ref.=1 0.654
## 2 0.433
## 3 0.679
## 4 0.932
## 6 0.254
## 7 0.877
##
## educ: ref.=1 0.564
## 2 0.566
## 3 0.148
## 4 0.483
## 5 0.273
##
## masts: ref.=1 0.194
## 2 0.513
## 3 0.045
## 4 0.834
## 5 0.312
## 6 0.469
##
## finc (cont. var.) 0.102 0.098
##
## PIR (cont. var.) 0.614 0.614
##
## SMQ020: 2 vs 1 0.492 0.492
##
## ALQ101: 2 vs 1 0.377 0.374
##
## PAQ665: 2 vs 1 0.004 0.003
##
## PAQ650: 2 vs 1 0.14 0.127
##
## BMI (cont. var.) < 0.001 < 0.001
##
## SBP (cont. var.) 0.214 0.211
##
## DBP (cont. var.) 0.982 0.982
##
## AST (cont. var.) 0.787 0.791
##
## ALT (cont. var.) 0.53 0.537
##
## BUN (cont. var.) 0.505 0.504
##
## crea (cont. var.) 0.522 0.563
##
## TC (cont. var.) 0.558 0.558
##
## TG (cont. var.) 0.509 0.509
##
## HDL (cont. var.) 0.535 0.535
##
## LDL (cont. var.) 0.561 0.561
##
## apoB (cont. var.) 0.444 0.44
##
## hyper: 1 vs 0 0.051 0.051
##
## db: 1 vs 0 0.483 0.485
##
## dyslip: 1 vs 0 0.451 0.454
##
## Log-likelihood = -496.313
## No. of observations = 1930
## AIC value = 1070.6261mo_sCOT = glm(dp10 ~ sCOT + age + sex + educ + masts + PIR + SMQ020 + ALQ101 + PAQ665 + BMI + SBP + DBP + AST + ALT + Glu + BUN + crea + TC + TG + HDL + LDL + apoB + hyper + db + dyslip, family = binomial, data = dat)
logistic.display(mo_sCOT)##
## Logistic regression predicting dp10 : 1 vs 0
##
## crude OR(95%CI) adj. OR(95%CI)
## sCOT (cont. var.) 1.0015 (1.0006,1.0024) 1.0011 (1,1.0022)
##
## age (cont. var.) 1.0152 (1.0061,1.0244) 1.0062 (0.9908,1.0217)
##
## sex: 2 vs 1 1.74 (1.26,2.41) 1.81 (1.2,2.73)
##
## educ: ref.=1
## 2 0.66 (0.36,1.19) 0.91 (0.47,1.77)
## 3 0.45 (0.25,0.8) 0.67 (0.35,1.28)
## 4 0.52 (0.3,0.89) 0.88 (0.47,1.65)
## 5 0.2 (0.11,0.38) 0.67 (0.31,1.43)
##
## masts: ref.=1
## 2 2.33 (1.37,3.96) 1.38 (0.74,2.58)
## 3 2.72 (1.77,4.18) 1.83 (1.15,2.93)
## 4 2.42 (1.15,5.13) 1.25 (0.54,2.91)
## 5 0.87 (0.53,1.42) 0.81 (0.47,1.43)
## 6 1.23 (0.65,2.33) 0.88 (0.44,1.77)
##
## PIR (cont. var.) 0.69 (0.61,0.77) 0.74 (0.64,0.85)
##
## SMQ020: 2 vs 1 0.59 (0.43,0.81) 0.84 (0.57,1.23)
##
## ALQ101: 2 vs 1 1.03 (0.73,1.46) 0.8 (0.53,1.21)
##
## PAQ665: 2 vs 1 2.43 (1.7,3.48) 1.83 (1.25,2.69)
##
## BMI (cont. var.) 1.06 (1.04,1.08) 1.05 (1.02,1.07)
##
## SBP (cont. var.) 1.0088 (1.0004,1.0172) 0.9933 (0.9821,1.0046)
##
## DBP (cont. var.) 1.001 (0.9887,1.0135) 0.9996 (0.9853,1.0141)
##
## AST (cont. var.) 1.0026 (0.9985,1.0066) 1.0017 (0.9947,1.0088)
##
## ALT (cont. var.) 1.005 (0.998,1.0121) 1.0027 (0.9906,1.0149)
##
## Glu (cont. var.) 1.0073 (1.0037,1.0109) 1.0022 (0.9968,1.0075)
##
## BUN (cont. var.) 1.0254 (1.0017,1.0497) 1.0089 (0.9772,1.0416)
##
## crea (cont. var.) 1.07 (0.85,1.35) 1.15 (0.82,1.62)
##
## TC (cont. var.) 1 (1,1.01) 0.86 (0.48,1.54)
##
## TG (cont. var.) 1.01 (1,1.01) 1.04 (0.92,1.16)
##
## HDL (cont. var.) 0.99 (0.98,1) 1.18 (0.66,2.11)
##
## LDL (cont. var.) 1 (1,1.01) 1.17 (0.65,2.09)
##
## apoB (cont. var.) 1.009 (1.0029,1.0152) 1.0062 (0.9862,1.0265)
##
## hyper: 1 vs 0 2.24 (1.63,3.07) 1.51 (0.98,2.33)
##
## db: 1 vs 0 1.89 (1.33,2.69) 1.02 (0.61,1.7)
##
## dyslip: 1 vs 0 1.54 (1,2.37) 0.83 (0.48,1.44)
##
## P(Wald's test) P(LR-test)
## sCOT (cont. var.) 0.043 0.046
##
## age (cont. var.) 0.433 0.433
##
## sex: 2 vs 1 0.005 0.004
##
## educ: ref.=1 0.565
## 2 0.792
## 3 0.226
## 4 0.698
## 5 0.301
##
## masts: ref.=1 0.128
## 2 0.305
## 3 0.011
## 4 0.597
## 5 0.473
## 6 0.717
##
## PIR (cont. var.) < 0.001 < 0.001
##
## SMQ020: 2 vs 1 0.36 0.359
##
## ALQ101: 2 vs 1 0.289 0.285
##
## PAQ665: 2 vs 1 0.002 0.001
##
## BMI (cont. var.) < 0.001 < 0.001
##
## SBP (cont. var.) 0.245 0.242
##
## DBP (cont. var.) 0.956 0.956
##
## AST (cont. var.) 0.628 0.639
##
## ALT (cont. var.) 0.663 0.666
##
## Glu (cont. var.) 0.426 0.432
##
## BUN (cont. var.) 0.586 0.586
##
## crea (cont. var.) 0.426 0.481
##
## TC (cont. var.) 0.604 0.603
##
## TG (cont. var.) 0.553 0.553
##
## HDL (cont. var.) 0.583 0.583
##
## LDL (cont. var.) 0.605 0.605
##
## apoB (cont. var.) 0.548 0.545
##
## hyper: 1 vs 0 0.061 0.06
##
## db: 1 vs 0 0.948 0.948
##
## dyslip: 1 vs 0 0.507 0.51
##
## Log-likelihood = -500.4239
## No. of observations = 1928
## AIC value = 1066.8478# Figure of OR (95% CI) of having Depressive Disorder;
ggcoef(mo_sCOT, exponentiate=T, exclude_intercept=T, vline_color = "red", errorbar_color = "blue", errorbar_height = 0.10)
# Danh gia tinh phan dinh (Discrimination)
library(Epi)
ROC(form = dp10 ~ sCOT + age + sex + educ + masts + PIR + SMQ020 + ALQ101 + PAQ665 + BMI + SBP + DBP + AST + ALT + Glu + BUN + crea + TC + TG + HDL + LDL + apoB + hyper + db + dyslip, data=dat, plot = "ROC", lwd=2) 