Table 5 Model performance of three regression methods for mapping the PHQ-8 to the HUI3 utility scores

1

BETAMIX M1a

1 component without truncation

Probability mass at full health

−0.2680

0.2844

0.3644

17

17

17

2

BETAMIX M1b

2 components without truncation

Probability mass at full health

−0.0013

0.1664

0.2077

13

13

13

3

BETAMIX M1c

2 components with truncation

Probability mass at full health

−0.0001

0.1666

0.2082

14

14

14

4

BETAMIX M1d

2 components with truncation

Probability mass at full health and truncation point

−0.0007

0.1662

0.2074

12

12

12

5

BETAMIX M2a

1 component without truncation

Probability mass at full health

−0.2682

0.2842

0.3624

16

16

16

6

BETAMIX M2b

2 components without truncation

Probability mass at full health

.

.

.

7

BETAMIX M2c

2 components with truncation

Probability mass at full health

−0.0010

0.1607

0.2023

7

1

4

8

BETAMIX M2d

2 components with truncation

Probability mass at full health and truncation point

0.0002

0.1606

0.2025

6

6

6

9

BETAMIX M3a

1 component without truncation

Probability mass at full health

−0.2679

0.2839

0.3623

15

15

15

10

BETAMIX M3b

2 components without truncation

Probability mass at full health

−0.0048

0.1602

0.2027

5

7

6

11

BETAMIX M3c

2 components with truncation

Probability mass at full health

−0.0026

0.1612

0.2028

8

8

8

12

BETAMIX M3d

2 components with truncation

Probability mass at full health and truncation point

−0.0158

0.1634

0.2048

11

9

10

13

OLS M1

0.0000

0.1630

0.2066

10

10

10

14

OLS M2

0.0000

0.1584

0.2024

3

3

3

15

OLS M3

0.0000

0.1584

0.2024

4

2

3

16

TOBIT M1

−0.0011

0.1627

0.2066

9

11

10

17

TOBIT M2

−0.0012

0.1579

0.2024

2

4

3

18

TOBIT M3

−0.0012

0.1579

0.2024

1

5

3