Table 1.
Characteristics of SIBDCS study participants
| Variables at enrolment (validation data) Median (1st quartile-3rd quartile); min–max |
SIBDCSa training data subset (N = 663) | SIBDCS validation data subset (N = 456) |
|||
|---|---|---|---|---|---|
| Low NEO-FFI risk (N = 221) |
High NEO-FFI risk (N = 235) |
P value | |||
| NEO-FFI risk groups (Cut-off = 11.3)b | High risk | 319/663 (48.1%) | – | – | – |
| NEO-FFI risk score | Across groups | – |
11.5 (8.8–14.1); 2.7–22.1 |
– | |
| 11.1 (8.8–13.4); 0.0–24.5 | 8.8 (7.4–9.9); 2.7–11.3 | 14.0 (12.5–15.9); 11.3–22.1 | By definition- | ||
| Type D personality | Yes | 192/654 (29.4%) | 29/219 (13.2%) | 113/228 (49.6%) | < 0.001 |
| Diagnosis |
CD UC/IBD-U |
380/663 (57.3%) | 116/221 (52.5%) | 125/235 (53.2%) | 0.995 |
| 283/663 (42.7%) | 105/221 (47.5%) | 110/235 (46.8%) | |||
| Sex | Male | 327/663 (49.3%) | 127/221 (57.5%) | 80/235 (34.0%) | < 0.001 |
| Age (years) | 41.6 (31.0–52.6); 16.5–84.6 | 39.6 (30.4–53.0); 17.8–81.8 | 41.9 (31.3–51.9); 14.8–80.6 | 0.746 | |
| Time since diagnosis (years) | 7.9 (3.0–15.9); 0.1–52.5 | 7.1 (2.1–15.3); 0.0–43.4 | 9.7 (2.8–18.3); 0.1–48.7 | 0.042 | |
| BMI (kg/m2) | 23.1 (20.9–26.0); 14.7–49.5 | 23.8 (21.6–26.1); 17.3–55.1 | 22.8 (20.7–26.3); 15.8–44.8 | 0.053 | |
| Smoking status | Smoker | 185/659 (28.1%) | 53/219 (24.2%) | 74/232 (31.9%) | 0.087 |
| Daily alcohol consumption | Yes | 47/655 (7.2%) | 24/217 (11.1%) | 17/231 (7.4%) | 0.233 |
| Disease activity | CDAI (CD patients) | 34.0 (13.0–73.3); 0.0–226.0 | 34.0 (16.0–78.5); 0.0–198.0 | 53.0 (25.0–106.0); − 0.0 to 450.0 | 0.007 |
| MTWAI (UC/IBD-U patients) | 2.5 (1.0–5.0); 0.0–16.0 | 2.0 (1.0–4.5); 0.0–19.0 | 3.0 (2.0–6.0); 0.0–18.0 | 0.049 | |
| Complications at or prior to enrolment | Yes | 217/508 (42.7%) | 78/174 (44.8%) | 97/184 (52.7%) | 0.166 |
| Current treatment with systemic steroids | Yes | 152/505 (30.1%) | 32/171 (18.7%) | 48/183 (26.2%) | 0.049 |
| Current treatment with TNF inhibitors | Yes | 85/505 (16.8%) | 37/171 (21.6%) | 43/183 (23.5%) | 0.771 |
| Number current therapies | 1.0 (1.0–2.0); 0.0–6.0 | 2 (1–2); 0–5 | 2 (1–2); 0–7 | 0.076 | |
| Disease-related surgery prior to enrolment | Yes | 175/508 (34.4%) | 55/174 (31.6%) | 67/184 (36.4%) | 0.397 |
| Fistula at or prior to enrolment | Yes | 141/508 (27.8%) | 45/174 (25.9%) | 52/184 (28.3%) | 0.700 |
| Stenosis at or prior to enrolment | Yes | 107/508 (21.1%) | 34/174 (19.5%) | 52/184 (28.3%) | 0.071 |
| EIM at or prior to enrolment | Yes | 182/508 (35.8%) | 68/174 (39.1%) | 88/184 (47.8%) | 0.119 |
| Depressive symptoms (HADS-D ≥ 11) | Yes | 45/655 (6.9%) | 5/220 (2.3%) | 30/231 (13.0%) | < 0.001 |
| IBDQ score | 182.0 (156.0–200.0);53.0–223.0 | 194.5 (171.0–206.0);87.0–222.0 | 169.0 (143.2–190.0);68.0–224.0 | < 0.001 | |
| Low IBDQ (IBDQ < 170) | Yes | 229/652 (35.1%) | 49/220 (22.3%) | 118/230 (51.3%) | < 0.001 |
| Marital status | Married | 327/639 (52.2%) | 108/217 (49.8%) | 119/225 (52.9%) | 0.575 |
| Academic education | Yes | 92/646 (14.2%) | 39/219 (17.8%) | 26/226 (11.5%) | 0.080 |
| Physically active | Yes | 411/640 (64.2%) | 143/212 (67.5%) | 153/224 (68.3%) | 0.930 |
| Observation time (years) for different endpoints | Active diseasec | 8.2 (6.1–9.6); 0.8–11.2 | 8.3 (6.1–9.7); 1.1–11.2 | 8.8 (7.2–9.8); 1.0–11.1 | 0.152 |
| Any new EIMc | 8.1 (6.1–9.5); 0.8–11.2 | 8.2 (6.1–9.6); 1.1–11.1 | 8.5 (7.1–9.8); 1.0–11.1 | 0.115 | |
| FNCEc | 5.1 (3.0–7.2); 0.4–11.2 | 5.1 (2.9–7.6); 0.8–10.7 | 5.5 (2.8–7.8); 0.8–11.1 | 0.361 | |
| AFFSSTc | 5.2 (3.3–7.5); 0.5–11.2 | 5.5 (3.1–7.6); 0.9–11.2 | 5.8 (3.1–7.5); 0.76–11.1 | 0.931 | |
| Depressive symptoms (HADS-D ≥ 11) | 5.1 (3.3–6.6); 0.6–8.1 | 5.4 (3.8–6.5); 0.8–9.4 | 5.0 (3.0–6.2); 0.7–8.0 | 0.143 | |
| Low IBDQ (IBDQ < 170) | 5.1 (3.2–6.6); 0.6–8.1 | 5.4 (3.8–6.6); 0.8–9.4 | 5.0 (3.0–6.2); 0.7–8.0 | 0.074 | |
P values were calculated using the Mann–Whitney U or χ2 tests, respectively and refer to differences between NEO-FFI risk score groups observed within the validation data subset. When characteristics of the training and validation subset were compared, higher CDAI scores (median: training data = 34.0 vs. validation data = 45.0, p = 0.011), more EIM at or prior to enrolment (training data = 35.8% vs. validation data = 43.6%, p = 0.026) and more individuals with current anti-TNF treatment (training data = 16.8% vs. validation data = 22.6%, p = 0.043) in the validation data subset were observed. No other significant differences were detected. Observation numbers diverging from 1119 for the whole SIBDCS are due to missing data for certain variables
CD Crohn’s disease, UC ulcerative colitis, IBD-U IBD unclassified, BMI body mass index, CDAI Crohn’s disease activity index, EIM extraintestinal manifestation, MTWAI modified truelove and witts severity Index, NEO-FFI NEO-five-factor inventory, TNF tumour necrosis factor, HADS-D hospital anxiety and depression scale subscale for depressive symptoms
a1154 SIBDCS patients for which at least 1 NEO-FFI cluster subcomponent had been assessed were included. However, only the 1119 patients with data for all 3 cluster subcomponents (i.e., self-reproach, negative affect and activity) on which the NEO-FFI risk score is based are presented here
bThe cut-off value (11.3) to interpret NEO-FFI risk groups is the mean NEO-FFI risk score of all included SIBDCS patients (training and validation data)
cPrecise definitions of the respective clinical endpoint can be found in the Methods
Bold values indicate significant P values (i.e. P < 0.05)