Skip to main content

Table 1 Descriptive statistics of predictors and outcome in the development and validation datasets

From: Multiclass risk models for ovarian malignancy: an illustration of prediction uncertainty due to the choice of algorithm

Characteristic

Development

External validation

Number of patients

5909

3199

Number of centers

24

25

Number of oncology centers

13a

15

Age (years)

47 (35–60)

49 (36–62)

Serum CA125 (U/mL)

26 (13–109)

25 (11–109)

Missing CA-125

1805 (31%)

966 (30%)

Maximal diameter of lesion (mm)

69 (48–100)

71 (50–105)

Proportion of solid tissue

0.11 (0-0.66)

0.06 (0-0.67)

Papillary projections

  

 0

4771 (81%)

2711 (85%)

 1

495 (8%)

200 (6%)

 2

148 (3%)

76 (2%)

 3

137 (2%)

51 (2%)

 > 3

358 (6%)

161 (5%)

> 10 cyst locules

471 (8%)

311 (12%)

Ascites

720 (12%)

321 (10%)

Shadows

742 (13%)

464 (14%)

Tumor outcome

  

 Benign

3980 (67%)

1988 (62%)

 Borderline

339 (6%)

259 (8%)

 Stage I invasive

356 (6%)

219 (7%)

 Stage II-IV invasive

988 (17%)

561 (18%)

 Secondary metastasis

246 (4%)

172 (5%)

  1. Results are shown as median (interquartile range) for continuous variables, and as n (%) for categorical variables
  2. a One center changed from a non-oncology center to an oncology center during the study period (Bologna, Italy). As a result, there were 24 centers yet 13 oncology and 12 non-oncology centers