Table 3 Performance Statistics of Models

Palliative Prognostic Index (PPI)

Cheng et al., 2012 [26]

3 weeks^a

6

71

68

81

56

0.68

Maltoni et al., 2012 [28]

30 days^a

6

73.7 (68.4–79)

67.1 (61.7–72.6)

67.8 (62.4–73.2)

73.1 (67.7–78.5)

0.62 (0.60–0.65)

Hung, 2014 [30]

30 days^a

8

58.9

64.8

73.7

48.4

0.66 (0.63—0.69)

Kao et al., 2014 [31]

30 days

5

79.5

50.8

57.1

75

0.63 (0.61–0.65)

Kim et al., 2014 [32]

3 weeks

5

60

63.3

45.4

75.7

0.65 (0.61—0.70)

C. Palomar-Munoz et al., 2018 [40]

3 weeks^a

6

79

51

66

66

N/A

Ermacora et al., 2018 [41]

30 days

N/A

N/A

0.72 (0.67–0.77)

Miyagi et al., 2020 [46]

3 weeks

N/A

N/A

0.76 (0.64–0.88)

Hiratsuka et al., 2022_c [51]

30 days^a

N/A

N/A

0.74 (0.72—0.76)

Combination of initial palliative prognostic Index (PPI) and week 1 PPI

Kao et al., 2014 [31]

30 days

4

66.9

77

70.6

73.8

0.71 (0.69—0.73)

Survival Prediction Score (SPS): 3-variable model

Chow et al., 2008 [21]

N/A

N/A

N/A

0.63

Number of risk factors (NRF): 3-variable model

Chow et al., 2008 [21]

N/A

N/A

N/A

0.63

A proposed prognostic 7-day survival formula

Chiang et al., 2009 [23]

1 week

0.2

71

75.7

26.8

90.1

N/A

Recursive partitioning: 2-variable model

Chow et al., 2009 [24]

N/A

N/A

N/A

0.61

Survival Prediction Score (SPS): 6-variable model

Chow et al., 2009 [22]

N/A

N/A

N/A

0.65

Number of risk factors (NRF): 6-variable model

Chow et al., 2009 [22]

N/A

N/A

N/A

0.65

Palliative Prognostic Score (PaP)

Scarpi et al., 2011 [25]

30 days

N/A

N/A

Maltoni et al., 2012 [28]

30 days^a

9

69.9 (64.4–75.4)

83.7 (79.3–88.2)

80.2 (75.0–85.3)

74.8 (70.0–79.5)

0.72 (0.70–0.73)

Kim et al., 2014 [32]

3 weels

10

72.9

74.2

59

84.3

0.81 (0.77—0.85)

[42]

30-days

N/A

N/A

0.87 (0.85—0.89)

Ermacora et al., 2018 [41]

30 days

N/A

N/A

0.82 (0.77–0.86)

Miyagi et al., 2020 [46]

3 weeks

N/A

N/A

0.86 (0.79–0.93)

Hiratsuka et al., 2022_a [49]

30 days

N/A

N/A

Japan = 0.75 (0.73–0.78),

Korea = 0.66 (0.6—0.72), Taiwan = 0.67 (0.61—0.74)

Hiratsuka et al., 2022_b [50]

30 days^a

N/A

91.1 (88.9–92.9)

40.2 (36.1–44.4)

68.8 (67.3–70.4)

75.6 (70.8–79.8)

Japan = 0.70 (0.68—0.73)

Korea = 0.71 (0.64—0.77)

Hiratsuka et al., 2022_c [51]

30 days^a

N/A

N/A

0.84 (0.82—0.86)

R. Mendis et al., 2015 [36]

30 days

N/A

N/A

0.71 (0.68–0.74)

Modified Palliative Prognostic Score—Delirium (D-PaP)

Hamano et al., 2018 [25]

30 days

N/A

N/A

N/A

Maltoni et al., 2012 [28]

3 weeks

9

72.9 (67.6–78.3)

80.2 (75.6–84.9)

77.6 (72.4–82.8)

75.9 (71.1–80.8)

76.7 (72.7—80.7)

Palliative Prognostic Score—Nomogram (PaP-Nomogram)

Scarpi et al., 2022 [54]

15-days^a

Various survival probability based on nomogram points

N/A

0.74 (0.72—0.75)

Cochin Risk Index Score (CRIS)

Durand et al., 2012 [27]

2 week

7

70

62

78

N/A

Palliative Performance Scale (PPS)

Maltoni et al., 2012 [28]

3 weeks^a

60

N/A

0.63 (0.60–0.66)

Kim et al., 2014 [32]

3 weeks ^a

30

65

69.8

52.3

79.7

0.729 (0.68—0.77)

Hiratsuka et al., 2022_c [51]

30 days ^a

Not specified

N/A

0.73 (0.70—0.75)

Prognostic Scale for terminal hospitalized chinese cancer patients (8-variable)

Huang et al., 2014 [29]

30 days

4

70

77

78

73

N/A

A graphic tool to estimate individualized survival curves (5-variable)

Chiang et al., 2015 [35]

Analysis by survival curve only

N/A

N/A

0.69

PRONOPALL score (4-variables)

Bourgeois etal., 2017 [38]

2 months^a

N/A

89.4

60.9

41.2

76.9

0.81 (0.75—0.87)

Objective Prognostic Score (OPS)

Yoon et al., 2014 [33]

3 week

3

83.6

56.8

77.8

65.6

0.74

Yoon et al., 2017 [38]

3-week

3

73.6

66.2

79.8

58

0.74 (0.68—0.81)

Ermacora et al., 2018 [41]

30 days

N/A

N/A

0.70 (0.64–0.75)

Hiratsuka et al., 2022_b [50]

30 days^a

3

43.6 (40.1–47.1)

87.8 (84.7–90.4)

83.8 (80.3–86.7)

51.7 (50.0–53.5

Japan: 0.70 (0.68—0.73)

Korea: 0.71 (0.64—0.77)

Imminent Mortality Predictor for Advanced Cancer (IMPAC)

Adelson et al., 2018 [39]

90-days^a

50%

40

N/A

60

N/A

0.72

Objective Prognostic Index for advanced cancer (OPI-AC) (7-days)

Hamano et al., 2018 [42]

7-days ^a

N/A

N/A

0.77 (0.66—0.87)

Prognosis in Palliative Care study (PiPS-B14/56)

Hamano et al., 2018 [42]

14-days ^a

N/A

N/A

0.86 (0.84—0.89)

Six adaptable prognosis prediction (SAP) model

Hamano et al., 2018[42]

30-days ^a

N/A

N/A

0.74 (0.65—0.83)

Nomogram based parameters to predict 90-days survival

Zhao et al., 2019 [43]

90 days

N/A

N/A

0.75 (0.70—0.80)

Artifical Neural network for 30-days survival prediction

Arkin et al., 2020 [44]

30-days

N/A

38

100

N/A

N/A

0.86

Logistic regression for 30-days survival

Arkin et al., 2020 [44]

30-days

N/A

48

84

N/A

N/A

0.76

Prognostic model for advanced cancer (PRO-MAC)

Hum et al., 2020 [45]

30-days ^a

4

66.9

68.1

57.1

76.5

0.73 (0.69–0.75)

Supportive and Palliative Care indicator tools

Chan et al., 2022 [48]

6 months

N/A

83.5

61

66.4

80

N/A

Rothman Index

Chan et al., 2022 [48]

6 months

60

69.7

11.9

42.2

29.8

N/A

Patient-Generated Subjective Global Assessment Short form (PG-SGA SF)

Cunha et al., 2022

90 days

15

60.2

70.1

N/A

N/A

0.75 (0.67—0.80)

Modified Barretos Prognostic Nomogram (BPN)—with laboratory values

Preto et al., 2022 [53]

30-days^a

N/A

N/A

0.78 (0.74—0.81)

Modified Barretos Prognostic Nomogram (BPN)—without laboratory values

Preto et al., 2022 [53]

30-days^a

N/A

N/A

0.74 (0.71—0.77)

Machine learning (Gradient-boosted trees binary classifier)

Zachariah et al., 2022 [55]

90-days

N/A

29.5

N/A

60

N/A

0.81 (0.83—0.91)

Objective Palliative Prognostic Score

Chen et al., 2015 [34]

1 week

3 out of 6 variables reached

68.8

86

55.9

91.4

0.82 (0.75—0.89)

Clinical Model

Owusuaa et al., 2022 [52]

1-year

40%

80

69

65

83

0.76 (0.73–0.78)

Extended Model

Owusuaa et al., 2022 [52]

1-year

40%

76

72

66

81

0.78 (0.76–0.80)

Data mining techniques (random forest algorithms, support-vector machine algorithms, back-propagation neural network algorithms)

Yang et al., 2021 [47]^b

Classification into < 30 days, 30–90 and > 90 days

N/A

N/A

N/A