Skip to main content

Table 3 Estimated cumulative mortality risks per 100 people associated with hypothetical interventions on physical activity (PA) and Mediterranean diet score (MDS)

From: Sustained adherence to a Mediterranean diet and physical activity on all-cause mortality in the Melbourne Collaborative Cohort Study: application of the g-formula

 

Pre-T2 mortality risk (95% CI)c

Post-T2 mortality risk (95% CI)c

Totala mortality risk (95% CI)c

Risk differenceb

(95% CI)c

Risk Ratiob

(95% CI)c

PAF

(95% CI)c

Observed risks

 Among whole cohort

4.58

(4.31, 4.86)

5.40

(5.10, 5.71)

9.74

(9.35, 10.13)

      

 Among followed-upd

4.58

(4.31, 4.86)

4.16

(3.79, 4.53)

8.55

(8.12, 8.99)

      

Estimated risks under no intervention

 Natural course

4.59

(4.30, 4.87)

5.54

(5.22, 5.87)

9.88

(9.47, 10.28)

Ref

 

Ref

   

Estimated risks under hypothetical interventionse

Intervene on everyone

 Control interventions (T1 and T2)

  Low PA & low MDS

5.68

(4.87, 6.50)

7.66

(6.38, 8.94)

12.90

(11.54, 14.27)

3.03

(1.75, 4.30)

1.31

(1.18, 1.44)

−0.31

(−0.44, − 0.18)

  Medium PA & MDS

4.53

(3.87, 5.19)

4.66

(3.79, 5.52)

8.97

(7.97, 9.98)

−0.90

(−1.86, 0.06)

0.91

(0.81, 1.01)

0.09

(−0.01, 0.19)

  Low PA

5.15

(4.63, 5.68)

7.10

(6.18, 8.03)

11.89

(10.90, 12.88)

2.01

(1.13, 2.89)

1.20

(1.11, 1.29)

−0.20

(− 0.29, − 0.11)

  Low MDS

4.63

(4.08, 5.17)

6.39

(5.68, 7.10)

10.72

(9.88, 11.55)

0.84

(0.15, 1.53)

1.09

(1.02, 1.15)

− 0.09

(− 0.15, − 0.02)

 Single interventions (T1)

  High PA

4.35

(3.69, 5.00)

4.94

(4.19, 5.70)

9.08

(8.12, 10.03)

−0.80

(−1.67, 0.07)

0.92

(0.83, 1.01)

0.08

(−0.01, 0.17)

  High MDS

4.71

(3.94, 5.49)

5.18

(4.44, 5.93)

9.65

(8.67, 10.63)

−0.22

(−1.13, 0.68)

0.98

(0.89, 1.07)

0.02

(−0.07, 0.11)

  High PA & MDS

4.63

(3.24, 6.03)

4.40

(3.04, 5.77)

8.83

(6.95, 10.71)

−1.05

(−2.90, 0.81)

0.89

(0.71, 1.08)

0.11

(−0.08, 0.29)

 Repeated interventions (T1 and T2)

 High PA

4.35

(3.69, 5.00)

4.39

(3.58, 5.19)

8.54

(7.56, 9.53)

−1.33

(−2.23, − 0.43)

0.87

(0.77, 0.96)

0.13

(0.04, 0.23)

  High MDS

4.71

(3.94, 5.48)

4.73

(3.93, 5.54)

9.22

(8.17, 10.27)

−0.65

(−1.64, 0.33)

0.93

(0.83, 1.03)

0.07

(−0.03, 0.17)

  High PA & MDS

4.63

(3.24, 6.03)

3.59

(2.22, 4.96)

8.06

(6.25, 9.87)

−1.82

(−3.60, −0.03)

0.82

(0.64, 1.00)

0.18

(0.00, 0.36)

Intervene on obese only

 Control interventions (T1 and T2)

  Low PA & MDS

4.73

(4.42, 5.03)

5.86

(5.48, 6.24)

10.31

(9.85, 10.77)

0.43

(0.24, 0.62)

1.04

(1.02, 1.06)

  

  Medium PA & MDS

4.56

(4.26, 4.87)

5.33

(4.99, 5.67)

9.65

(9.23, 10.08)

−0.22

(−0.40, −0.05)

0.98

(0.96, 1.00)

  

  Low PA

4.66

(4.36, 4.95)

5.75

(5.40, 6.09)

10.14

(9.71, 10.56)

0.26

(0.13, 0.39)

1.03

(1.01, 1.04)

  

 Low MDS

4.60

(4.31, 4.90)

5.67

(5.33, 6.02)

10.02

(9.59, 10.44)

0.14

(0.02, 0.26)

1.01

(1.00, 1.03)

  

Single interventions (T1)

  High PA

4.54

(4.25, 4.83)

5.45

(5.11, 5.78)

9.74

(9.31, 10.17)

−0.14

(−0.30, 0.02)

0.99

(0.97, 1.00)

  

  High MDS

4.60

(4.29, 4.90)

5.51

(5.16, 5.86)

9.85

(9.42, 10.29)

−0.02

(− 0.19, 0.14)

1.00

(0.98, 1.01)

  

  High PA & MDS

4.58

(4.23, 4.92)

5.36

(5.00, 5.73)

9.70

(9.22, 10.17)

−0.18

(−0.46, 0.10)

0.98

(0.95, 1.01)

  

 Repeated interventions (T1 and T2)

  High PA

4.54

(4.25, 4.83)

5.29

(4.95, 5.64)

9.59

(9.17, 10.02)

−0.28

(−0.45, − 0.12)

0.97

(0.95, 0.99)

  

  High MDS

4.60

(4.28, 4.91)

5.39

(5.04, 5.75)

9.74

(9.30, 10.18)

−0.14

(−0.33, 0.06)

0.99

(0.97, 1.01)

  

  High PA & MDS

4.58

(4.23, 4.93)

5.15

(4.78, 5.52)

9.49

(9.01, 9.97)

−0.38

(−0.67, − 0.10)

0.96

(0.93, 0.99)

  
  1. PA Physical activity, MDS Mediterranean Diet Score, PAF population attributable fraction, CI confidence interval. T1 is study baseline is (1995–1999), T0 is initial cohort (MCCS) recruitment (1990–1994), T2 is follow-up (2003–2007). a Estimated as: Pr (Dead) = Pr (Dead pre-T2) + Pr (Dead post-T2|Alive at T2) x Pr (Alive at T2); b Compared with natural course; c 95% normal-based bootstrap confidence intervals using 100 bootstrap samples per imputed dataset; d Death post-T2 estimated using participants followed-up at T2 – the estimator we would obtain if outcome data were available only on followed-up participants, as is often the case. eMDS groups: low (scores 0–3), medium (4–5) and high (6–9); Physical activity groups: low (e.g. walking twice a week or less), medium (e.g. jogging twice a week), high (e.g. running three or more times a week)
Back to article page

BMC Public Health

ISSN: 1471-2458

Contact us