Skip to main content

Table 2 Breakdown of the number of individuals with missing data

From: How are missing data in covariates handled in observational time-to-event studies in oncology? A systematic review

Description

Number

(%)

Excluded missing data in initial phase (N=106)

  

Excluded individuals with missing data in any covariate1

44

(42)

Excluded individuals with missing data in a subset of covariates

62

(58)

Reported the number of individuals excluded

66

(62)

Percentage (%) of individuals excluded (n=66)

  

Mean (SD)

14.14

(12.40)

Median (IQR)

10.22

(4.73, 18.34)

Min, Max

0.11

47.38

Missing data present for the analysis stage (N=102)

  

Reported missing data in baseline table for incomplete covariates

82

(80)

Used a complete-case analysis2

35

(34)

Used other missing data methods

36

(35)

Quantified the complete-case sample size

25

(25)

Percentage (%) of individuals excluded (n=25)

  

Mean (SD)

31.65

(21.90)

Median (IQR)

31.34

(13.67, 37.76)

Min, Max

1.77

94.16

  1. The initial phase is the stage when defining the study population using inclusion and exclusion criteria.
  2. 1potentially used a complete-case in initial phase but did not clearly state their methods
  3. 2A further 31 were not clear on whether they used a complete-case during the analysis