Introduce jstable

Jinseob Kim

2019-11-08

Regression Tables from ‘GLM’, ‘GEE’, ‘GLMM’, ‘Cox’ and ‘survey’ Results for Publication.

Install

remotes::install_github('jinseob2kim/jstable')
library(jstable)

GLM Table

## Gaussian
glm_gaussian <- glm(mpg~cyl + disp, data = mtcars)
glmshow.display(glm_gaussian, decimal = 2)
#> $first.line
#> [1] "Linear regression predicting mpg\n"
#> 
#> $table
#>      crude coeff.(95%CI)   crude P value adj. coeff.(95%CI)   
#> cyl  "-2.88 (-3.51,-2.24)" "< 0.001"     "-1.59 (-2.98,-0.19)"
#> disp "-0.04 (-0.05,-0.03)" "< 0.001"     "-0.02 (-0.04,0)"    
#>      adj. P value
#> cyl  "0.034"     
#> disp "0.054"     
#> 
#> $last.lines
#> [1] "No. of observations = 32\nR-squared = 0.7596\nAIC value = 167.1456\n\n"
#> 
#> attr(,"class")
#> [1] "display" "list"

## Binomial
glm_binomial <- glm(vs~cyl + disp, data = mtcars, family = binomial)
glmshow.display(glm_binomial, decimal = 2)
#> $first.line
#> [1] "Logistic regression predicting vs\n"
#> 
#> $table
#>      crude OR.(95%CI)   crude P value adj. OR.(95%CI)    adj. P value
#> cyl  "0.2 (0.08,0.56)"  "0.002"       "0.15 (0.02,1.02)" "0.053"     
#> disp "0.98 (0.97,0.99)" "0.002"       "1 (0.98,1.03)"    "0.715"     
#> 
#> $last.lines
#> [1] "No. of observations = 32\nAIC value = 23.8304\n\n"
#> 
#> attr(,"class")
#> [1] "display" "list"

GEE Table: from geeglm object from geepack package

library(geepack)  ## for dietox data
data(dietox)
dietox$Cu <- as.factor(dietox$Cu)
dietox$ddn = as.numeric(rnorm(nrow(dietox)) > 0)
gee01 <- geeglm (Weight ~ Time + Cu , id = Pig, data = dietox, family = gaussian, corstr = "ex")
geeglm.display(gee01)
#> $caption
#> [1] "GEE(gaussian) predicting Weight by Time, Cu - Group Pig"
#> 
#> $table
#>            crude coeff(95%CI)   crude P value adj. coeff(95%CI)  
#> Time       "6.94 (6.79,7.1)"    "< 0.001"     "6.94 (6.79,7.1)"  
#> Cu: ref.=1 NA                   NA            NA                 
#>    2       "-0.59 (-3.73,2.54)" "0.711"       "-0.84 (-3.9,2.23)"
#>    3       "1.9 (-1.87,5.66)"   "0.324"       "1.77 (-1.9,5.45)" 
#>            adj. P value
#> Time       "< 0.001"   
#> Cu: ref.=1 NA          
#>    2       "0.593"     
#>    3       "0.345"     
#> 
#> $metric
#>                                  crude coeff(95%CI) crude P value
#>                                  NA                 NA           
#> Estimated correlation parameters "0.775"            NA           
#> No. of clusters                  "72"               NA           
#> No. of observations              "861"              NA           
#>                                  adj. coeff(95%CI) adj. P value
#>                                  NA                NA          
#> Estimated correlation parameters NA                NA          
#> No. of clusters                  NA                NA          
#> No. of observations              NA                NA

gee02 <- geeglm (ddn ~ Time + Cu , id = Pig, data = dietox, family = binomial, corstr = "ex")
geeglm.display(gee02)
#> $caption
#> [1] "GEE(binomial) predicting ddn by Time, Cu - Group Pig"
#> 
#> $table
#>            crude OR(95%CI)    crude P value adj. OR(95%CI)    
#> Time       "0.98 (0.95,1.02)" "0.331"       "0.98 (0.95,1.02)"
#> Cu: ref.=1 NA                 NA            NA                
#>    2       "0.88 (0.63,1.22)" "0.436"       "0.88 (0.63,1.22)"
#>    3       "0.93 (0.67,1.3)"  "0.682"       "0.93 (0.67,1.3)" 
#>            adj. P value
#> Time       "0.333"     
#> Cu: ref.=1 NA          
#>    2       "0.438"     
#>    3       "0.684"     
#> 
#> $metric
#>                                  crude OR(95%CI) crude P value
#>                                  NA              NA           
#> Estimated correlation parameters "0.003"         NA           
#> No. of clusters                  "72"            NA           
#> No. of observations              "861"           NA           
#>                                  adj. OR(95%CI) adj. P value
#>                                  NA             NA          
#> Estimated correlation parameters NA             NA          
#> No. of clusters                  NA             NA          
#> No. of observations              NA             NA

Mixed model Table: lmerMod or glmerMod object from lme4 package

library(lme4)
l1 = lmer(Weight ~ Time + Cu + (1|Pig), data = dietox) 
lmer.display(l1, ci.ranef = T)
#> $table
#>                      crude coeff(95%CI) crude P value adj. coeff(95%CI)
#> Time                   6.94 (6.88,7.01)     0.0000000  6.94 (6.88,7.01)
#> Cu: ref.=1                         <NA>            NA              <NA>
#>    2                 -0.58 (-4.67,3.51)     0.7811327 -0.84 (-4.47,2.8)
#>    3                   1.9 (-2.23,6.04)     0.3670740  1.77 (-1.9,5.45)
#> Random effects                     <NA>            NA              <NA>
#> Pig                 40.34 (28.08,54.95)            NA              <NA>
#> Residual             11.37 (10.3,12.55)            NA              <NA>
#> Metrics                            <NA>            NA              <NA>
#> No. of groups (Pig)                  72            NA              <NA>
#> No. of observations                 861            NA              <NA>
#> Log-likelihood                  -2400.8            NA              <NA>
#> AIC value                        4801.6            NA              <NA>
#>                     adj. P value
#> Time                   0.0000000
#> Cu: ref.=1                    NA
#>    2                   0.6527264
#>    3                   0.3442310
#> Random effects                NA
#> Pig                           NA
#> Residual                      NA
#> Metrics                       NA
#> No. of groups (Pig)           NA
#> No. of observations           NA
#> Log-likelihood                NA
#> AIC value                     NA
#> 
#> $caption
#> [1] "Linear mixed model fit by REML : Weight ~ Time + Cu + (1 | Pig)"

l2 = glmer(ddn ~  Time + Cu + (1|Pig), data= dietox, family= "binomial")
lmer.display(l2)
#> $table
#>                      crude OR(95%CI) crude P value   adj. OR(95%CI)
#> Time                0.98 (0.94,1.02)     0.3720006 0.98 (0.94,1.02)
#> Cu: ref.=1                      <NA>            NA             <NA>
#>    2                0.88 (0.63,1.22)     0.4342310 0.88 (0.63,1.22)
#>    3                0.93 (0.67,1.31)     0.6853029 0.93 (0.67,1.31)
#> Random effects                  <NA>            NA             <NA>
#> Pig                             0.01            NA             <NA>
#> Metrics                         <NA>            NA             <NA>
#> No. of groups (Pig)               72            NA             <NA>
#> No. of observations              861            NA             <NA>
#> Log-likelihood               -595.44            NA             <NA>
#> AIC value                    1200.87            NA             <NA>
#>                     adj. P value
#> Time                   0.3740683
#> Cu: ref.=1                    NA
#>    2                   0.4368170
#>    3                   0.6869323
#> Random effects                NA
#> Pig                           NA
#> Metrics                       NA
#> No. of groups (Pig)           NA
#> No. of observations           NA
#> Log-likelihood                NA
#> AIC value                     NA
#> 
#> $caption
#> [1] "Generalized linear mixed model fit by maximum likelihood (Laplace Approximation) : ddn ~ Time + Cu + (1 | Pig)"

Cox model with frailty or cluster options

library(survival)
fit1 <- coxph(Surv(time, status) ~ ph.ecog + age, cluster = inst, lung, model = T)  ## model = T: to extract original data
fit2 <- coxph(Surv(time, status) ~ ph.ecog + age + frailty(inst), lung, model = T)
cox2.display(fit1)
#> $table
#>         crude HR(95%CI)    crude P value adj. HR(95%CI)  adj. P value
#> ph.ecog "1.61 (1.25,2.08)" "< 0.001"     "1.56 (1.22,2)" "< 0.001"   
#> age     "1.02 (1,1.03)"    "0.008"       "1.01 (1,1.02)" "0.085"     
#> 
#> $ranef
#>         [,1] [,2] [,3] [,4]
#> cluster   NA   NA   NA   NA
#> inst      NA   NA   NA   NA
#> 
#> $metric
#>                     [,1] [,2] [,3] [,4]
#> <NA>                  NA   NA   NA   NA
#> No. of observations  226   NA   NA   NA
#> No. of events        163   NA   NA   NA
#> 
#> $caption
#> [1] "Marginal Cox model on time ('time') to event ('status') - Group inst"
cox2.display(fit2)
#> $table
#>         crude HR(95%CI)    crude P value adj. HR(95%CI)     adj. P value
#> ph.ecog "1.64 (1.31,2.05)" "< 0.001"     "1.58 (1.26,1.99)" "< 0.001"   
#> age     "1.02 (1,1.04)"    "0.043"       "1.01 (0.99,1.03)" "0.225"     
#> 
#> $ranef
#>         [,1] [,2] [,3] [,4]
#> frailty   NA   NA   NA   NA
#> inst      NA   NA   NA   NA
#> 
#> $metric
#>                     [,1] [,2] [,3] [,4]
#> <NA>                  NA   NA   NA   NA
#> No. of observations  226   NA   NA   NA
#> No. of events        163   NA   NA   NA
#> 
#> $caption
#> [1] "Frailty Cox model on time ('time') to event ('status') - Group inst"

Cox mixed effect model Table: coxme object from coxme package

library(coxme)
fit <- coxme(Surv(time, status) ~ ph.ecog + age + (1|inst), lung)
coxme.display(fit) 
#> $table
#>         crude HR(95%CI)    crude P value adj. HR(95%CI)     adj. P value
#> ph.ecog "1.66 (1.32,2.09)" "< 0.001"     "1.61 (1.27,2.03)" "< 0.001"   
#> age     "1.02 (1,1.04)"    "0.043"       "1.01 (0.99,1.03)" "0.227"     
#> 
#> $ranef
#>                 [,1] [,2] [,3] [,4]
#> Random effect     NA   NA   NA   NA
#> inst(Intercept) 0.02   NA   NA   NA
#> 
#> $metric
#>                     [,1] [,2] [,3] [,4]
#> <NA>                  NA   NA   NA   NA
#> No. of groups(inst)   18   NA   NA   NA
#> No. of observations  226   NA   NA   NA
#> No. of events        163   NA   NA   NA
#> 
#> $caption
#> [1] "Mixed effects Cox model on time ('time') to event ('status') - Group inst"

GLM for survey data : svyglm object from survey package

library(survey)
data(api)
apistrat$tt = c(rep(1, 20), rep(0, nrow(apistrat) -20))
apistrat$tt2 = factor(c(rep(0, 40), rep(1, nrow(apistrat) -40)))

dstrat<-svydesign(id=~1,strata=~stype, weights=~pw, data=apistrat, fpc=~fpc)
ds <- svyglm(api00~ell+meals + tt2, design=dstrat)
ds2 <- svyglm(tt~ell+meals+ tt2, design=dstrat, family = quasibinomial())
svyregress.display(ds)
#> $first.line
#> [1] "Linear regression predicting api00- weighted data\n"
#> 
#> $table
#>             crude coeff.(95%CI)    crude P value adj. coeff.(95%CI)   
#> ell         "-3.73 (-4.35,-3.11)"  "< 0.001"     "-0.51 (-1.27,0.26)" 
#> meals       "-3.38 (-3.71,-3.05)"  "< 0.001"     "-3.11 (-3.65,-2.57)"
#> tt2: 1 vs 0 "10.98 (-34.16,56.12)" "0.634"       "6.24 (-17.68,30.17)"
#>             adj. P value
#> ell         "0.195"     
#> meals       "< 0.001"   
#> tt2: 1 vs 0 "0.61"      
#> 
#> $last.lines
#> [1] "No. of observations = 200\nAIC value = 2308.0628\n\n"
#> 
#> attr(,"class")
#> [1] "display" "list"
svyregress.display(ds2)
#> $first.line
#> [1] "Logistic regression predicting tt- weighted data\n"
#> 
#> $table
#>             crude OR.(95%CI)   crude P value adj. OR.(95%CI)   
#> ell         "1.02 (1,1.05)"    "0.047"       "1.11 (1.02,1.21)"
#> meals       "1.01 (0.99,1.03)" "0.255"       "0.97 (0.93,1.01)"
#> tt2: 1 vs 0 "0 (0,0)"          "< 0.001"     "0 (0,0)"         
#>             adj. P value
#> ell         "0.02"      
#> meals       "0.151"     
#> tt2: 1 vs 0 "< 0.001"   
#> 
#> $last.lines
#> [1] "No. of observations = 200\n\n"
#> 
#> attr(,"class")
#> [1] "display" "list"

Cox model for survey data :svycoxph object from survey package

data(pbc, package="survival")
pbc$sex = factor(pbc$sex)
pbc$stage = factor(pbc$stage)
pbc$randomized<-with(pbc, !is.na(trt) & trt>0)
biasmodel<-glm(randomized~age*edema,data=pbc,family=binomial)
pbc$randprob<-fitted(biasmodel)

if (is.null(pbc$albumin)) pbc$albumin<-pbc$alb ##pre2.9.0

dpbc <- svydesign(id=~1, prob=~randprob, strata=~edema, data=subset(pbc,randomized))

model <- svycoxph(Surv(time,status>0)~ sex + protime + albumin + stage,design=dpbc)
svycox.display(model)
#> Stratified Independent Sampling design (with replacement)
#> svydesign(id = ~1, prob = ~randprob, strata = ~edema, data = subset(pbc, 
#>     randomized))
#> Stratified Independent Sampling design (with replacement)
#> svydesign(id = ~1, prob = ~randprob, strata = ~edema, data = subset(pbc, 
#>     randomized))
#> Stratified Independent Sampling design (with replacement)
#> svydesign(id = ~1, prob = ~randprob, strata = ~edema, data = subset(pbc, 
#>     randomized))
#> Stratified Independent Sampling design (with replacement)
#> svydesign(id = ~1, prob = ~randprob, strata = ~edema, data = subset(pbc, 
#>     randomized))
#> Stratified Independent Sampling design (with replacement)
#> svydesign(id = ~1, prob = ~randprob, strata = ~edema, data = subset(pbc, 
#>     randomized))
#> $table
#>               crude HR(95%CI)      crude P value adj. HR(95%CI)       
#> sex: f vs m   "0.62 (0.4,0.97)"    "0.038"       "0.55 (0.33,0.9)"    
#> protime       "1.37 (1.09,1.72)"   "0.006"       "1.52 (1.2,1.91)"    
#> albumin       "0.2 (0.14,0.29)"    "< 0.001"     "0.31 (0.2,0.47)"    
#> stage: ref.=1 NA                   NA            NA                   
#>    2          "5.67 (0.77,41.78)"  "0.089"       "10.94 (1.01,118.55)"
#>    3          "9.78 (1.37,69.94)"  "0.023"       "17.03 (1.69,171.6)" 
#>    4          "22.89 (3.2,163.48)" "0.002"       "22.56 (2.25,226.42)"
#>               adj. P value
#> sex: f vs m   "0.017"     
#> protime       "< 0.001"   
#> albumin       "< 0.001"   
#> stage: ref.=1 NA          
#>    2          "0.049"     
#>    3          "0.016"     
#>    4          "0.008"     
#> 
#> $metric
#>                        [,1] [,2] [,3] [,4]
#> <NA>                     NA   NA   NA   NA
#> No. of observations  312.00   NA   NA   NA
#> No. of events        144.00   NA   NA   NA
#> AIC                 1483.12   NA   NA   NA
#> 
#> $caption
#> [1] "Survey cox model on time ('time') to event ('status > 0')"

Sub-group analysis for Cox/svycox model

library(dplyr)
lung %>% 
mutate(status = as.integer(status == 1),
       sex = factor(sex),
       kk = factor(as.integer(pat.karno >= 70)),
       kk1 = factor(as.integer(pat.karno >= 60))) -> lung

#TableSubgroupMultiCox(Surv(time, status) ~ sex, var_subgroups = c("kk", "kk1"), data=lung, line = T)

## Survey data
library(survey)
data.design <- svydesign(id = ~1, data = lung, weights = ~1)
#TableSubgroupMultiCox(Surv(time, status) ~ sex, var_subgroups = c("kk", "kk1"), data = data.design, line = F)