SG_Study_SII_SIRI_analysis
Stephen Murata
11/29/2021
#Summary & missingness tables
createTable(compareGroups(~., data=SG_df_new))##
## --------Summary descriptives table ---------
##
## _______________________________
## [ALL] N
## N=79
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
## Pt_Group: 79
## HC 32 (40.5%)
## Patient 47 (59.5%)
## Exp_Group: 47
## ESC_PBO 21 (44.7%)
## ESC_CBX 26 (55.3%)
## Ethnicity: 46
## 1 30 (65.2%)
## 2 10 (21.7%)
## 3 5 (10.9%)
## 5 1 (2.17%)
## BMI 31.6 (5.78) 45
## Age 42.0 (12.7) 46
## Sex: 46
## 1 16 (34.8%)
## 2 30 (65.2%)
## SII_BL 489 (230) 79
## SII_WK8 462 (219) 47
## SIRI_BL 1.03 (0.60) 79
## SIRI_WK8 1.07 (0.61) 47
## HAMD17_BL 22.5 (6.34) 46
## HAMD17_WK8 10.0 (6.32) 43
## Remission: 43
## Non-Remitter 25 (58.1%)
## Remitter 18 (41.9%)
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯missingTable(compareGroups(Exp_Group~., data=SG_df_new))##
## --------Missingness table by 'Exp_Group'---------
##
## ________________________________________
## ESC_PBO ESC_CBX p.overall
## N=21 N=26
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
## Subject_ID 0 (0.00%) 0 (0.00%) .
## Pt_Group 0 (0.00%) 0 (0.00%) .
## Ethnicity 1 (4.76%) 0 (0.00%) 0.447
## BMI 2 (9.52%) 0 (0.00%) 0.194
## Age 1 (4.76%) 0 (0.00%) 0.447
## Sex 1 (4.76%) 0 (0.00%) 0.447
## SII_BL 0 (0.00%) 0 (0.00%) .
## SII_WK8 0 (0.00%) 0 (0.00%) .
## SIRI_BL 0 (0.00%) 0 (0.00%) .
## SIRI_WK8 0 (0.00%) 0 (0.00%) .
## HAMD17_BL 1 (4.76%) 0 (0.00%) 0.447
## HAMD17_WK8 3 (14.3%) 1 (3.85%) 0.311
## Remission 3 (14.3%) 1 (3.85%) 0.311
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯1 Outcome inspection (HAMD17)
mu <- ddply(SG_df_new_long, "Timepoint", summarise, grp.mean=mean(HAMD17))
ggplot(SG_df_new_long, aes(x=HAMD17))+
geom_histogram(color="black", fill="orange")+
facet_grid(Timepoint ~ .)+
theme(legend.position="none")+
geom_vline(data=mu, aes(xintercept=grp.mean, color=Timepoint),linetype="dashed")+
labs(title="Distribution of HAMD17 by treatment timepoint", x="HAMD17", y="Count")+
theme_gray()## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
ggplot(SG_df_new_long, aes(x = Timepoint, y = HAMD17))+
geom_boxplot(aes(fill=Timepoint))+
geom_jitter(width = 0.1)+
facet_wrap(~Treatment)+
theme_bw()+
theme(legend.position = "none")+
stat_compare_means(method="t.test")
2 Predictor variable inspection (SII)
mu <- ddply(SG_df_new_long, "Treatment", summarise, grp.mean=mean(SII))
ggplot(SG_df_new_long, aes(x=SII))+
geom_histogram(color="black", fill="orange")+
facet_grid(Treatment ~ .)+
theme(legend.position="none")+
geom_vline(data=mu, aes(xintercept=grp.mean, color=Treatment),linetype="dashed")+
labs(title="Distribution of SII by treatment arm", x="SII", y="Count")+
theme_gray()## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
mu <- ddply(SG_df_new_long, "Timepoint", summarise, grp.mean=mean(SII))
ggplot(SG_df_new_long, aes(x=SII))+
geom_histogram(color="black", fill="orange")+
facet_grid(Timepoint ~ .)+
theme(legend.position="none")+
geom_vline(data=mu, aes(xintercept=grp.mean, color=Timepoint),linetype="dashed")+
labs(title="Distribution of SII by timepoint", x="SII", y="Count")+
theme_gray()## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
ggplot(SG_df_new_long, aes(x = Timepoint, y = SII))+
geom_boxplot(aes(fill=Timepoint))+
geom_jitter(width = 0.1)+
facet_wrap(~Treatment)+
theme_bw()+
theme(legend.position = "none")+
stat_compare_means(method="t.test")
3 Predictor variable inspection (SII)
mu <- ddply(SG_df_new_long, "Treatment", summarise, grp.mean=mean(SIRI))
ggplot(SG_df_new_long, aes(x=SIRI))+
geom_histogram(color="black", fill="orange")+
facet_grid(Treatment ~ .)+
theme(legend.position="none")+
geom_vline(data=mu, aes(xintercept=grp.mean, color=Treatment),linetype="dashed")+
labs(title="Distribution of SIRI by treatment arm", x="SIRI", y="Count")+
theme_gray()## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
mu <- ddply(SG_df_new_long, "Timepoint", summarise, grp.mean=mean(SIRI))
ggplot(SG_df_new_long, aes(x=SIRI))+
geom_histogram(color="black", fill="orange")+
facet_grid(Timepoint ~ .)+
theme(legend.position="none")+
geom_vline(data=mu, aes(xintercept=grp.mean, color=Timepoint),linetype="dashed")+
labs(title="Distribution of SIRI by timepoint", x="SIRI", y="Count")+
theme_gray()## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
ggplot(SG_df_new_long, aes(x = Timepoint, y = SIRI))+
geom_boxplot(aes(fill=Timepoint))+
geom_jitter(width = 0.1)+
facet_wrap(~Treatment)+
theme_bw()+
theme(legend.position = "none")+
stat_compare_means(method="t.test")
# Group comparison by patient status
createTable(compareGroups
(Pt_Group ~ .,
data = SG_df_new,
method = NA),
hide.no = '0',
show.p.mul= T,
show.all = TRUE,
show.ratio = TRUE)##
## --------Summary descriptives table by 'Pt_Group'---------
##
## ______________________________________________________________________________________________________
## [ALL] HC Patient OR p.ratio p.overall
## N=79 N=32 N=47
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
## Exp_Group: .
## ESC_PBO 21 (44.7%) 0 (.%) 21 (44.7%) Ref. Ref.
## ESC_CBX 26 (55.3%) 0 (.%) 26 (55.3%) . [.;.] .
## Ethnicity: .
## 1 30 (65.2%) 0 (.%) 30 (65.2%) Ref. Ref.
## 2 10 (21.7%) 0 (.%) 10 (21.7%) . [.;.] .
## 3 5 (10.9%) 0 (.%) 5 (10.9%) . [.;.] .
## 5 1 (2.17%) 0 (.%) 1 (2.17%) . [.;.] .
## BMI 31.6 (5.78) . (.) 31.6 (5.78) . [.;.] . .
## Age 42.0 (12.7) . (.) 42.0 (12.7) . [.;.] . .
## Sex: .
## 1 16 (34.8%) 0 (.%) 16 (34.8%) Ref. Ref.
## 2 30 (65.2%) 0 (.%) 30 (65.2%) . [.;.] .
## SII_BL 477 [310;627] 491 [305;611] 477 [320;640] 1.00 [1.00;1.00] 0.530 0.834
## SII_WK8 408 [309;579] . [.;.] 408 [309;579] . [.;.] . .
## SIRI_BL 0.85 [0.60;1.27] 0.76 [0.58;1.21] 0.98 [0.64;1.39] 1.82 [0.80;4.17] 0.154 0.201
## SIRI_WK8 1.02 [0.62;1.38] . [.;.] 1.02 [0.62;1.38] . [.;.] . .
## HAMD17_BL 22.5 (6.34) . (.) 22.5 (6.34) . [.;.] . .
## HAMD17_WK8 9.00 [6.00;14.0] . [.;.] 9.00 [6.00;14.0] . [.;.] . .
## Remission: .
## Non-Remitter 25 (58.1%) 0 (.%) 25 (58.1%) Ref. Ref.
## Remitter 18 (41.9%) 0 (.%) 18 (41.9%) . [.;.] .
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯Take home message: SII and SIRI levels were similar in TRBDD compared to HC’s
4 Group comparison by treatment arm
##
## --------Summary descriptives table by 'Exp_Group'---------
##
## _______________________________________________________________________________________________________
## [ALL] ESC_PBO ESC_CBX OR p.ratio p.overall
## N=47 N=21 N=26
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
## Pt_Group: Patient 47 (100%) 21 (100%) 26 (100%) Ref. Ref. .
## Ethnicity: 0.658
## 1 30 (65.2%) 12 (60.0%) 18 (69.2%) Ref. Ref.
## 2 10 (21.7%) 4 (20.0%) 6 (23.1%) . [.;.] .
## 3 5 (10.9%) 3 (15.0%) 2 (7.69%) . [.;.] .
## 5 1 (2.17%) 1 (5.00%) 0 (0.00%) . [.;.] .
## BMI 31.6 (5.78) 32.2 (4.89) 31.1 (6.40) 0.97 [0.87;1.07] 0.524 0.515
## Age 42.0 (12.7) 46.7 (13.0) 38.3 (11.3) 0.94 [0.90;0.99] 0.032 0.028
## Sex: 0.776
## 1 16 (34.8%) 6 (30.0%) 10 (38.5%) Ref. Ref.
## 2 30 (65.2%) 14 (70.0%) 16 (61.5%) 0.70 [0.19;2.42] 0.572
## SII_BL 477 [320;640] 523 [357;759] 379 [296;618] 1.00 [1.00;1.00] 0.137 0.168
## SII_WK8 408 [309;579] 399 [320;639] 415 [297;518] 1.00 [1.00;1.00] 0.196 0.386
## SIRI_BL 0.98 [0.64;1.39] 1.06 [0.78;1.53] 0.95 [0.59;1.25] 0.60 [0.24;1.51] 0.277 0.294
## SIRI_WK8 1.02 [0.62;1.38] 1.04 [0.73;1.69] 0.94 [0.57;1.22] 0.38 [0.13;1.11] 0.076 0.118
## HAMD17_BL 22.5 (6.34) 21.6 (7.31) 23.1 (5.56) 1.04 [0.94;1.14] 0.447 0.472
## HAMD17_WK8 9.00 [6.00;14.0] 11.5 [9.00;16.0] 7.00 [5.00;11.0] 0.81 [0.70;0.94] 0.007 0.001
## Remission: <0.001
## Non-Remitter 25 (58.1%) 17 (94.4%) 8 (32.0%) Ref. Ref.
## Remitter 18 (41.9%) 1 (5.56%) 17 (68.0%) 29.6 [4.73;800] <0.001
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯Take home message: patients in ESC_CBX arm were significantly younger, had more remitters, and lower HAMD17_WK8
5 Summary of sample by remission
createTable(compareGroups
(Remission ~ .,
data = SG_df_new,
method = NA),
hide.no = '0',
show.p.mul= T,
show.all = TRUE,
show.ratio = TRUE)##
## --------Summary descriptives table by 'Remission'---------
##
## _______________________________________________________________________________________________________
## [ALL] Non-Remitter Remitter OR p.ratio p.overall
## N=43 N=25 N=18
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
## Pt_Group: Patient 43 (100%) 25 (100%) 18 (100%) Ref. Ref. .
## Exp_Group: <0.001
## ESC_PBO 18 (41.9%) 17 (68.0%) 1 (5.56%) Ref. Ref.
## ESC_CBX 25 (58.1%) 8 (32.0%) 17 (94.4%) 29.6 [4.73;800] <0.001
## Ethnicity: 0.057
## 1 27 (64.3%) 15 (62.5%) 12 (66.7%) Ref. Ref.
## 2 9 (21.4%) 3 (12.5%) 6 (33.3%) . [.;.] .
## 3 5 (11.9%) 5 (20.8%) 0 (0.00%) . [.;.] .
## 5 1 (2.38%) 1 (4.17%) 0 (0.00%) . [.;.] .
## BMI 31.4 (5.82) 32.0 (4.85) 30.7 (7.00) 0.96 [0.86;1.07] 0.471 0.503
## Age 40.0 [34.0;49.2] 44.0 [34.8;53.8] 37.5 [31.8;43.2] 0.97 [0.92;1.02] 0.226 0.186
## Sex: 1.000
## 1 16 (38.1%) 9 (37.5%) 7 (38.9%) Ref. Ref.
## 2 26 (61.9%) 15 (62.5%) 11 (61.1%) 0.94 [0.26;3.46] 0.928
## SII_BL 443 [320;633] 494 [336;646] 436 [296;612] 1.00 [1.00;1.00] 0.570 0.649
## SII_WK8 422 [300;606] 451 [307;639] 407 [302;539] 1.00 [1.00;1.00] 0.343 0.453
## SIRI_BL 0.98 [0.68;1.34] 1.01 [0.78;1.38] 0.90 [0.48;1.17] 1.00 [0.39;2.53] 0.996 0.445
## SIRI_WK8 1.02 [0.66;1.42] 1.12 [0.69;1.69] 0.94 [0.52;1.23] 0.36 [0.11;1.18] 0.091 0.143
## HAMD17_BL 22.2 (5.87) 21.8 (5.86) 22.8 (6.00) 1.03 [0.93;1.14] 0.602 0.613
## HAMD17_WK8 9.00 [6.00;14.0] 13.0 [10.0;16.0] 5.50 [2.25;7.00] 0.00 [0.00;.] 0.998 <0.001
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯Take home message: clinical remission was significantly associated with ESC_CBX arm and lower HAMD17_WK8, but no other differences.
6 Group comparison by treatment timepoint
createTable(compareGroups
(Timepoint ~ .,
data = SG_df_new_long,
method = NA),
hide.no = '0',
show.ratio=TRUE,
show.p.mul= T)##
## --------Summary descriptives table by 'Timepoint'---------
##
## ________________________________________________________________________________
## Baseline Week 8 OR p.ratio p.overall
## N=79 N=47
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
## Pt_Group: <0.001
## HC 32 (40.5%) 0 (0.00%) Ref. Ref.
## TRBDD 47 (59.5%) 47 (100%) . [.;.] .
## Treatment: <0.001
## ESC_CBX 26 (32.9%) 26 (55.3%) Ref. Ref.
## ESC_PBO 21 (26.6%) 21 (44.7%) . [.;.] .
## HC 32 (40.5%) 0 (0.00%) . [.;.] .
## Ethnicity: 1.000
## 1 30 (65.2%) 30 (65.2%) Ref. Ref.
## 2 10 (21.7%) 10 (21.7%) 1.00 [0.36;2.82] 1.000
## 3 5 (10.9%) 5 (10.9%) 1.00 [0.25;4.08] 1.000
## 5 1 (2.17%) 1 (2.17%) 1.00 [0.02;40.2] 1.000
## BMI 30.7 [27.7;35.7] 30.7 [27.7;35.7] 1.00 [0.93;1.07] 1.000 1.000
## Age 40.0 [34.0;49.5] 40.0 [34.0;49.5] 1.00 [0.97;1.03] 1.000 1.000
## Sex: 1.000
## 1 16 (34.8%) 16 (34.8%) Ref. Ref.
## 2 30 (65.2%) 30 (65.2%) 1.00 [0.42;2.39] 1.000
## SII 477 [310;627] 408 [309;579] 1.00 [1.00;1.00] 0.516 0.446
## SIRI 0.85 [0.60;1.27] 1.02 [0.62;1.38] 1.12 [0.62;2.04] 0.707 0.639
## HAMD17 22.5 (6.34) 10.0 (6.32) 0.74 [0.66;0.84] <0.001 <0.001
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯createTable(compareGroups
(Timepoint ~ .,
data = SG_df_new_long,
method = NA,
subset = Treatment == "ESC_PBO"),
hide.no = '0',
show.ratio=TRUE,
show.p.mul= T)##
## --------Summary descriptives table by 'Timepoint'---------
##
## _______________________________________________________________________________________
## Baseline Week 8 OR p.ratio p.overall
## N=21 N=21
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
## Pt_Group: TRBDD 21 (100%) 21 (100%) Ref. Ref. .
## Treatment: ESC_PBO 21 (100%) 21 (100%) Ref. Ref. .
## Ethnicity: 1.000
## 1 12 (60.0%) 12 (60.0%) Ref. Ref.
## 2 4 (20.0%) 4 (20.0%) 1.00 [0.18;5.42] 1.000
## 3 3 (15.0%) 3 (15.0%) 1.00 [0.14;6.90] 1.000
## 5 1 (5.00%) 1 (5.00%) 1.00 [0.02;42.1] 1.000
## BMI 32.2 (4.89) 32.2 (4.89) 1.00 [0.87;1.14] 1.000 1.000
## Age 47.0 [35.0;59.5] 47.0 [35.0;59.5] 1.00 [0.95;1.05] 1.000 1.000
## Sex: 1.000
## 1 6 (30.0%) 6 (30.0%) Ref. Ref.
## 2 14 (70.0%) 14 (70.0%) 1.00 [0.25;4.06] 1.000
## SII 523 [357;759] 399 [320;639] 1.00 [1.00;1.00] 0.490 0.473
## SIRI 1.06 [0.78;1.53] 1.04 [0.73;1.69] 1.06 [0.44;2.53] 0.899 0.910
## HAMD17 22.0 [17.2;25.0] 11.5 [9.00;16.0] 0.84 [0.74;0.95] 0.005 0.001
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯createTable(compareGroups
(Timepoint ~ .,
data = SG_df_new_long,
method = NA,
subset = Treatment == "ESC_CBX"),
hide.no = '0',
show.ratio=TRUE,
show.p.mul= T)##
## --------Summary descriptives table by 'Timepoint'---------
##
## _______________________________________________________________________________________
## Baseline Week 8 OR p.ratio p.overall
## N=26 N=26
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
## Pt_Group: TRBDD 26 (100%) 26 (100%) Ref. Ref. .
## Treatment: ESC_CBX 26 (100%) 26 (100%) Ref. Ref. .
## Ethnicity: 1.000
## 1 18 (69.2%) 18 (69.2%) Ref. Ref.
## 2 6 (23.1%) 6 (23.1%) 1.00 [0.26;3.88] 1.000
## 3 2 (7.69%) 2 (7.69%) 1.00 [0.10;10.5] 1.000
## BMI 29.7 [27.0;35.2] 29.7 [27.0;35.2] 1.00 [0.92;1.09] 1.000 1.000
## Age 37.5 [31.0;43.2] 37.5 [31.0;43.2] 1.00 [0.95;1.05] 1.000 1.000
## Sex: 1.000
## 1 10 (38.5%) 10 (38.5%) Ref. Ref.
## 2 16 (61.5%) 16 (61.5%) 1.00 [0.32;3.13] 1.000
## SII 379 [296;618] 415 [297;518] 1.00 [1.00;1.00] 0.609 0.840
## SIRI 0.95 [0.59;1.25] 0.94 [0.57;1.22] 0.72 [0.25;2.04] 0.536 0.869
## HAMD17 23.1 (5.56) 7.52 (5.08) 0.48 [0.27;0.86] 0.013 <0.001
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯Take home message: no significant within-group differences by treatment timepoint
7 Reduced model: inspecting normality of residuals
Rcoef_list <- list()
beta_list<-list()
se_list<-list()
t_list<-list()
p_list<-list()
var_list<-list()
sg_bx_trans_vars<-SG_df_new %>% dplyr::select(contains("SII"), contains("SIRI")) %>% names() %>% sort()
for (x in sg_bx_trans_vars) {
LM1 <- lm(substitute(HAMD17_WK8 ~
Sex+
Age+
BMI+
Ethnicity+
Exp_Group+
HAMD17_BL+
i, list(i = as.name(x))), data = SG_df_new)
Rcoef_list[[x]]<- summary(LM1)$r.squared[1]
beta_list[[x]]<-summary(LM1)$coefficients[8,1]
se_list[[x]]<-summary(LM1)$coefficients[8,2]
t_list[[x]]<-summary(LM1)$coefficients[8,3]
p_list[[x]]<-summary(LM1)$coefficients[8,4]
var_list[[x]]<-x
plot(LM1, which=2, main=x)
}
do.call(rbind, Map(data.frame,
Beta=beta_list,
t_test=t_list,
SE=se_list,
p_value=p_list,
R_coeff=Rcoef_list)) %>%
knitr::kable(digits=2) | Beta | t_test | SE | p_value | R_coeff | |
|---|---|---|---|---|---|
| SII_BL | -5.03 | -2.70 | 1.86 | 0.01 | 0.47 |
| SII_WK8 | -4.34 | -2.35 | 1.85 | 0.03 | 0.47 |
| SIRI_BL | -5.06 | -2.69 | 1.88 | 0.01 | 0.46 |
| SIRI_WK8 | -4.24 | -2.19 | 1.93 | 0.04 | 0.46 |
8 Modelling HAMD17_WK8 by SII_BL
SII_model<-lm(HAMD17_WK8~Sex+Age+BMI+Ethnicity+Exp_Group+HAMD17_BL+SII_BL, data=SG_df_new)
summary(SII_model)##
## Call:
## lm(formula = HAMD17_WK8 ~ Sex + Age + BMI + Ethnicity + Exp_Group +
## HAMD17_BL + SII_BL, data = SG_df_new)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.8469 -3.0236 -0.4004 2.5441 12.5266
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 13.715142 7.239820 1.894 0.0672 .
## Sex2 -1.706205 1.967459 -0.867 0.3923
## Age 0.080664 0.074785 1.079 0.2888
## BMI -0.085822 0.156051 -0.550 0.5862
## Ethnicity2 -2.485789 2.186288 -1.137 0.2640
## Ethnicity3 0.450344 2.783369 0.162 0.8725
## Ethnicity5 15.939055 6.197756 2.572 0.0150 *
## Exp_GroupESC_CBX -5.028521 1.862032 -2.701 0.0110 *
## HAMD17_BL 0.064833 0.163512 0.397 0.6944
## SII_BL -0.003522 0.004201 -0.838 0.4081
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.29 on 32 degrees of freedom
## (37 observations deleted due to missingness)
## Multiple R-squared: 0.4666, Adjusted R-squared: 0.3166
## F-statistic: 3.111 on 9 and 32 DF, p-value: 0.008442SII_model<-lm(HAMD17_WK8~Exp_Group+HAMD17_BL+SII_BL, data=SG_df_new)
summary(SII_model)##
## Call:
## lm(formula = HAMD17_WK8 ~ Exp_Group + HAMD17_BL + SII_BL, data = SG_df_new)
##
## Residuals:
## Min 1Q Median 3Q Max
## -7.217 -4.391 -1.222 2.762 14.851
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 11.646889 4.398438 2.648 0.01163 *
## Exp_GroupESC_CBX -6.297679 1.820112 -3.460 0.00132 **
## HAMD17_BL 0.115995 0.153095 0.758 0.45320
## SII_BL -0.001051 0.003818 -0.275 0.78459
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.717 on 39 degrees of freedom
## (36 observations deleted due to missingness)
## Multiple R-squared: 0.2412, Adjusted R-squared: 0.1828
## F-statistic: 4.132 on 3 and 39 DF, p-value: 0.0123SII_model<-lm(HAMD17_WK8~Exp_Group+HAMD17_BL+SII_BL*Age, data=SG_df_new)
summary(SII_model)##
## Call:
## lm(formula = HAMD17_WK8 ~ Exp_Group + HAMD17_BL + SII_BL * Age,
## data = SG_df_new)
##
## Residuals:
## Min 1Q Median 3Q Max
## -9.1130 -2.9734 -0.8435 3.5895 9.1012
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 26.2085118 7.2287810 3.626 0.000885 ***
## Exp_GroupESC_CBX -4.6999379 1.6856282 -2.788 0.008412 **
## HAMD17_BL -0.0815423 0.1432295 -0.569 0.572680
## SII_BL -0.0342591 0.0098524 -3.477 0.001341 **
## Age -0.2795007 0.1267641 -2.205 0.033936 *
## SII_BL:Age 0.0008386 0.0002255 3.719 0.000678 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.886 on 36 degrees of freedom
## (37 observations deleted due to missingness)
## Multiple R-squared: 0.4882, Adjusted R-squared: 0.4171
## F-statistic: 6.868 on 5 and 36 DF, p-value: 0.0001365plot(SII_model, which=c(2,6))
#Subgroup analysis
mod_PBO<-lm(HAMD17_WK8~HAMD17_BL+SII_BL*Age, data=SG_df_new_PBO)
summary(mod_PBO)##
## Call:
## lm(formula = HAMD17_WK8 ~ HAMD17_BL + SII_BL * Age, data = SG_df_new_PBO)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.7275 -3.0296 -0.7293 3.7273 7.9895
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 24.7174303 14.6901538 1.683 0.1183
## HAMD17_BL -0.0102055 0.2280685 -0.045 0.9650
## SII_BL -0.0365582 0.0203096 -1.800 0.0970 .
## Age -0.3144870 0.2531744 -1.242 0.2379
## SII_BL:Age 0.0009503 0.0004174 2.277 0.0419 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.216 on 12 degrees of freedom
## (4 observations deleted due to missingness)
## Multiple R-squared: 0.5139, Adjusted R-squared: 0.3518
## F-statistic: 3.171 on 4 and 12 DF, p-value: 0.05388mod_CBX<-lm(HAMD17_WK8~HAMD17_BL+SII_BL*Age, data=SG_df_new_CBX)
summary(mod_CBX)##
## Call:
## lm(formula = HAMD17_WK8 ~ HAMD17_BL + SII_BL * Age, data = SG_df_new_CBX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.3116 -2.3769 -0.8257 3.6661 8.7459
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 21.0628303 8.6671997 2.430 0.0246 *
## HAMD17_BL -0.1103184 0.1993014 -0.554 0.5860
## SII_BL -0.0274848 0.0127549 -2.155 0.0435 *
## Age -0.2182576 0.1591229 -1.372 0.1854
## SII_BL:Age 0.0005816 0.0003307 1.759 0.0939 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.945 on 20 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.2114, Adjusted R-squared: 0.05368
## F-statistic: 1.34 on 4 and 20 DF, p-value: 0.2897Take home message: low depressive severity (post-treatment) is predicted by PBO arm, and lower SII (baseline) amongst pts with higher age
9 Modelling HAMD17_WK8 by SIRI_BL
SIRI_model<-lm(HAMD17_WK8~Sex+Age+BMI+Ethnicity+Exp_Group+HAMD17_BL+SIRI_BL, data=SG_df_new)
summary(SIRI_model)##
## Call:
## lm(formula = HAMD17_WK8 ~ Sex + Age + BMI + Ethnicity + Exp_Group +
## HAMD17_BL + SIRI_BL, data = SG_df_new)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.9137 -3.0266 -0.5317 2.6736 12.6592
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 13.90132 7.58632 1.832 0.0762 .
## Sex2 -1.99844 1.89570 -1.054 0.2997
## Age 0.09202 0.07307 1.259 0.2170
## BMI -0.09775 0.16057 -0.609 0.5470
## Ethnicity2 -2.88073 2.34326 -1.229 0.2279
## Ethnicity3 0.06538 2.80374 0.023 0.9815
## Ethnicity5 14.12885 5.89719 2.396 0.0226 *
## Exp_GroupESC_CBX -5.05550 1.88062 -2.688 0.0113 *
## HAMD17_BL 0.05112 0.16769 0.305 0.7624
## SIRI_BL -1.20602 1.55649 -0.775 0.4441
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.298 on 32 degrees of freedom
## (37 observations deleted due to missingness)
## Multiple R-squared: 0.4649, Adjusted R-squared: 0.3145
## F-statistic: 3.09 on 9 and 32 DF, p-value: 0.008781SIRI_model<-lm(HAMD17_WK8~Exp_Group+HAMD17_BL+SIRI_BL, data=SG_df_new)
summary(SIRI_model)##
## Call:
## lm(formula = HAMD17_WK8 ~ Exp_Group + HAMD17_BL + SIRI_BL, data = SG_df_new)
##
## Residuals:
## Min 1Q Median 3Q Max
## -7.297 -4.349 -1.190 2.801 14.518
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 11.7299 4.7338 2.478 0.01765 *
## Exp_GroupESC_CBX -6.2734 1.8075 -3.471 0.00128 **
## HAMD17_BL 0.1071 0.1629 0.658 0.51458
## SIRI_BL -0.3775 1.4767 -0.256 0.79959
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.718 on 39 degrees of freedom
## (36 observations deleted due to missingness)
## Multiple R-squared: 0.241, Adjusted R-squared: 0.1826
## F-statistic: 4.127 on 3 and 39 DF, p-value: 0.01236SIRI_model<-lm(HAMD17_WK8~Exp_Group+HAMD17_BL*SIRI_BL, data=SG_df_new)
summary(SIRI_model)##
## Call:
## lm(formula = HAMD17_WK8 ~ Exp_Group + HAMD17_BL * SIRI_BL, data = SG_df_new)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.492 -3.640 -1.241 2.689 13.015
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 32.3231 7.1674 4.510 6.06e-05 ***
## Exp_GroupESC_CBX -5.4085 1.6079 -3.364 0.00177 **
## HAMD17_BL -0.9779 0.3392 -2.883 0.00644 **
## SIRI_BL -20.2259 5.7722 -3.504 0.00119 **
## HAMD17_BL:SIRI_BL 1.0791 0.3058 3.529 0.00111 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.027 on 38 degrees of freedom
## (36 observations deleted due to missingness)
## Multiple R-squared: 0.4283, Adjusted R-squared: 0.3682
## F-statistic: 7.118 on 4 and 38 DF, p-value: 0.0002221plot(SIRI_model, which=c(2,6))
#subgroup analysis
mod_PBO<-lm(HAMD17_WK8~HAMD17_BL*SIRI_BL, data=SG_df_new_PBO)
summary(mod_PBO)##
## Call:
## lm(formula = HAMD17_WK8 ~ HAMD17_BL * SIRI_BL, data = SG_df_new_PBO)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.6504 -2.9974 -0.8465 2.4163 12.3178
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 39.1160 10.6675 3.667 0.00254 **
## HAMD17_BL -1.3944 0.5145 -2.710 0.01691 *
## SIRI_BL -27.5875 8.4249 -3.275 0.00554 **
## HAMD17_BL:SIRI_BL 1.5239 0.4469 3.410 0.00423 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.017 on 14 degrees of freedom
## (3 observations deleted due to missingness)
## Multiple R-squared: 0.4806, Adjusted R-squared: 0.3693
## F-statistic: 4.319 on 3 and 14 DF, p-value: 0.02363mod_CBX<-lm(HAMD17_WK8~HAMD17_BL*SIRI_BL, data=SG_df_new_CBX)
summary(mod_CBX)##
## Call:
## lm(formula = HAMD17_WK8 ~ HAMD17_BL * SIRI_BL, data = SG_df_new_CBX)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.4201 -3.4461 -0.8804 2.5984 8.6024
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 21.5205 9.9040 2.173 0.0414 *
## HAMD17_BL -0.6475 0.4591 -1.411 0.1730
## SIRI_BL -13.3246 8.1356 -1.638 0.1164
## HAMD17_BL:SIRI_BL 0.6569 0.4323 1.520 0.1435
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.103 on 21 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.1184, Adjusted R-squared: -0.007585
## F-statistic: 0.9398 on 3 and 21 DF, p-value: 0.4391Take home message: low depressive severity (post-treatment) is predicted by PBO arm, and also lower SIRI (baseline) amongst pts with higher baseline depression. The interaction effect appears to be carried mainly by PBO_ESC group.
10 Final model tables and graphics
tab_model(SII_model, SIRI_model)| Â | HAMD 17 WK 8 | HAMD 17 WK 8 | ||||
|---|---|---|---|---|---|---|
| Predictors | Estimates | CI | p | Estimates | CI | p |
| (Intercept) | 26.21 | 11.55 – 40.87 | 0.001 | 32.32 | 17.81 – 46.83 | <0.001 |
| Exp_Group [ESC_CBX] | -4.70 | -8.12 – -1.28 | 0.008 | -5.41 | -8.66 – -2.15 | 0.002 |
| HAMD17_BL | -0.08 | -0.37 – 0.21 | 0.573 | -0.98 | -1.66 – -0.29 | 0.006 |
| SII_BL | -0.03 | -0.05 – -0.01 | 0.001 | |||
| Age | -0.28 | -0.54 – -0.02 | 0.034 | |||
| SII_BL * Age | 0.00 | 0.00 – 0.00 | 0.001 | |||
| SIRI_BL | -20.23 | -31.91 – -8.54 | 0.001 | |||
| HAMD17_BL * SIRI_BL | 1.08 | 0.46 – 1.70 | 0.001 | |||
| Observations | 42 | 43 | ||||
| R2 / R2 adjusted | 0.488 / 0.417 | 0.428 / 0.368 | ||||
interact_plot(SII_model, pred = SII_BL, modx = Age, jitter=0.1, plot.points = TRUE, main.title = "The effect of SII_BL on HAMD17_WK8 depends on age")
interact_plot(SIRI_model, pred = SIRI_BL, modx = HAMD17_BL, jitter=0.1, plot.points = TRUE, main.title = "The effect of SIRI_BL on HAMD17_WK8 depends on HAMD17_BL")