Int J Sports Med 2009; 30(3): 213-224
DOI: 10.1055/s-0028-1128150
Training & Testing

© Georg Thieme Verlag KG Stuttgart · New York

Physical Activity and All-cause Mortality: An Updated Meta-analysis with Different Intensity Categories

H. Löllgen 1 , A. Böckenhoff 2 , G. Knapp 3
  • 1Department of Medicine, Ruhr-Universtity, Remscheid, Germany
  • 2Institute of Medical Informatics, University of Essen/Duisburg, Essen, Germany
  • 3Technological University Dortmund, Faculty of Statistics, Dortmund, Germany
Further Information

Publication History

accepted after revision December 9, 2008

Publication Date:
06 February 2009 (online)

Abstract

In a meta-analysis we investigated the effect of physical activity with different intensity categories on all-cause mortality. Many studies have reported positive effects of regular physical activity on primary prevention. This recent meta-analysis analyzed all-cause mortality with special reference to intensity categories. A computerized systematic literature search was performed in EMBASE, PUBMED, and MEDLINE data bases (1990–2006) for prospective cohort studies on physical leisure activity. Thirty-eight studies were identified and evaluated. The presentation refers to studies with 3 or 4 different intensities of regular physical activity according to a standard questionnaire. There was a significant association of lower all-cause mortality for active individuals compared with sedentary persons. For studies with three activity categories (mildly, moderately, and highly active) and multivariate-adjusted models, highly active men had a 22% lower risk of all-cause mortality (RR=0.78; 95% CI: 0.72 to 0.84) compared to mildly active men. For women, the relative risk was 0.69 (95% CI: 0.53 to 0.90). We observed similar results in moderately active persons compared to mildly active individuals (RR=0.81 for men and RR=0.76 for women). This association of activity to all-cause mortality was similar and significant in older subjects. Regular physical activity over longer time is strongly associated with a reduction in all-cause mortality in active subjects compared to sedentary persons. There is a dose-response curve especially from sedentary subjects to those with mild and moderate exercise with only a minor additional reduction with further increase in activity level.

References

Appendix A

Tables A1-A8

Table A1 Studies with two intensity groups of physical activity.

First author (year)

Country

Follow-up (yrs)

Sex

Adjustment

Age group

Number of participants

RR in the active group+(95% CI)

Kaplan (1996)

USA

28

all

multivariate

3

6 131

0.84 (0.77–0.92)

Knoops (2004)

Europe

10

all

multivariate

2

2 339

0.63 (0.55–0.72)

Hedblad (1997)

S

25

men

age

1

642

0.68 (0.49–0.94)

Hedblad (1997)

S

25

men

multivariate

1

642

0.70 (0.50–0.98)

Villeneuve (1998)

CDN

7

men

multivariate

1

6 246

0.82 (0.65–1.04)

Schnoor (2000)

DK

22

men

multivariate

3

4 658

0.39 (0.19–0.73)

Panagiotakos (2004)

GR

40

men

multivariate

1

529

0.83 (0.66–1.02)

Villeneuve (1998)

CDN

7

women

multivariate

1

8 196

0.88 (0.68–1.04)

Definition of age group: 1=”<65 years”, 2=”≥65 years”, 3=”all ages”

Table A2 Studies with three intensity groups of physical activity: Sex=all.

First author (year)

Country

Follow-up (yrs)

Adjustment

Age group

Number of participants

RR in the moderate active group+(95% CI)

RR in the most active group+(95% CI)

LaCroix (1996)

USA

5

age

2

1 645

0.69 (0.46–1.06)

0.73 (0.48–1.10)

Kujala (1998)

FIN

19

age

1

15 902

0.71 (0.62–0.81)

0.57 (0.45–0.74)

Hillsdon (2003)

GB

12

age

1

10 522

0.57 (0.42–0.79)

0.72 (0.54–0.95)

LaCroix (1996)

USA

5

multivariate

2

1 645

0.83 (0.53–1.29)

0.91 (0.58–1.42)

Kujala (1998)

FIN

19

multivariate

1

15 902

0.80 (0.69–0.91)

0.76 (0.59–0.98)

Hillsdon (2003)

GB

12

multivariate

1

10 522

0.63 (0.45–0.89)

0.81 (0.60–1.09)

Table A3 Studies with three intensity groups of physical activity: Sex=men.

First author (year)

Country

Follow-up (yrs)

Adjustment

Age group

Number of participants

RR in the moderate active group+(95% CI)

RR in the most active group+(95% CI)

Leon (1991)

USA

10.5

age

1

12 138

0.85 (0.73–0.99)

0.87 (0.74–1.01)

Paffenbarger (1993)

USA

9

age

3

10 269

0.70 (0.54–0.88)

0.68 (0.54–0.95)

LaCroix (1996)

USA

5

age

2

615

0.78 (0.43–1.45)

0.89 (0.49–1.62)

Morgan (1997)

GB

10

age

2

635

0.62 (0.38–1.00)

0.36 (0.26–0.51)

Rosengren (1997)

S

20

age

1

7 142

0.74 (0.68–0.82)

0.73 (0.68–0.79)

Bijnen (1998)

NL

10

age

2

802

0.67 (0.52–0.85)

0.64 (0.50–0.83)

Hakim (1998)

USA

12

age

2

707

0.64 (0.38–1.12)

0.53 (0.34–0.77)

Andersen (2000)

DK

14.5

age

3

17 265

0.71 (0.66–0.76)

0.65 (0.59–0.70)

Yu (2003)

GB

11

age

1

1 975

0.88 (0.66–1.18)

0.58 (0.41–0.82)

Barengo (2004)

FIN

20

age

1

15 853

0.85 (0.79–0.92)

0.60 (0.53–0.68)

Leon (1991)

USA

10.5

multivariate

1

12 138

0.89 (0.77–1.04)

0.92 (0.79–1.07)

Lindsted (1991)

USA

26

multivariate

1

9 484

0.68 (0.59–0.78)

0.76 (0.63–0.92)

Mensink (1996)

D

8

multivariate

1

954

0.61 (0.35–1.05)

0.79 (0.48–1.31)

Morgan (1997)

GB

10

multivariate

2

635

0.85 (0.52–1.39)

0.63 (0.44–0.89)

Rosengren (1997)

S

20

multivariate

1

7 142

0.84 (0.77–0.93)

0.83 (0.77–0.90)

Bijnen (1998)

NL

10

multivariate

2

802

0.80 (0.63–1.02)

0.77 (0.59–1.00)

Hakim (1998)

USA

12

multivariate

2

707

0.62 (0.36–1.08)

0.56 (0.37–0.83)

Sherman (1999)

USA

16

multivariate

3

962

0.85 (0.77–0.94)

0.80 (0.70–0.89)

Andersen (2000)

DK

14.5

multivariate

3

17 265

0.72 (0.66–0.78)

0.71 (0.65–0.78)

Yu (2003)

GB

11

multivariate

1

1 975

0.87 (0.65–1.17)

0.61 (0.43–0.86)

Barengo (2004)

FIN

20

multivariate

1

15 853

0.91 (0.85–0.98)

0.80 (0.71–0.90)

Table A4 Studies with three intensity groups of physical activity: Sex=women.

First author (year)

Country

Follow-up (yrs)

Adjustment

Age group

Number of participants

RR in the moderate active group+(95% CI)

RR in the most active group+(95% CI)

LaCroix (1996)

USA

5

age

2

1 030

0.50 (0.28–0.90)

0.45 (0.25–0.83)

Kushi (1997)

USA

7

age

1

40 417

0.66 (0.60–0.73)

0.58 (0.52–0.65)

Morgan (1997)

GB

10

age

2

635

0.61 (0.40–0.92)

0.36 (0.27–0.48)

Andersen (2000)

DK

14.5

age

3

13 375

0.64 (0.59–0.69)

0.55 (0.49–0.62)

Barengo (2004)

FIN

20

age

1

16 824

0.85 (0.77–0.94)

0.87 (0.74–1.02)

Lissner (1996)

S

20

multivariate

1

1 405

0.56 (0.39–0.82)

0.45 (0.24–0.86)

Mensink (1996)

D

8

multivariate

1

1 142

0.94 (0.51–1.75)

0.81 (0.44–1.49)

Kushi (1997)

USA

7

multivariate

1

40 417

0.77 (0.69–0.86)

0.68 (0.60–0.77)

Morgan (1997)

GB

10

multivariate

2

635

0.73 (0.48–1.12)

0.48 (0.36–0.65)

Sherman (1999)

USA

16

multivariate

3

1 410

0.84 (0.76–0.93)

0.83 (0.75–0.90)

Andersen (2000)

DK

14.5

multivariate

3

13 375

0.65 (0.60–0.71)

0.59 (0.52–0.67)

Barengo (2004)

FIN

20

multivariate

1

16 824

0.90 (0.82–0.99)

0.98 (0.83–1.15)

Table A5 Studies with four intensity groups of physical activity: Sex=all.

First author (year)

Country

Follow-up (yrs)

Adjustment

Age group

Number of participants

RR in the light active group+(95% CI)

RR in the moderate active group+(95% CI)

RR in the most active group+(95% CI)

Andersen (2000)

DK

14.5

age

3

30 640

0.68 (0.64–0.71)

0.61 (0.57–0.66)

0.53 (0.41–0.68)

Hillsdon (2003)

GB

12

age

1

7 704

0.68 (0.45–1.05)

0.40 (0.27–0.58)

0.40 (0.28–0.57)

Hillsdon (2003)

GB

12

multivariate

1

7 704

1.14 (0.74–1.78)

0.53 (0.35–0.82)

0.52 (0.35–0.78)

Table A6 Studies with four intensity groups of physical activity: Sex=men.

First author (year)

Country

Follow-up (yrs)

Adjustment

Age group

Number of participants

RR in the light active group+(95% CI)

RR in the moderate active group+(95% CI)

RR in the most active group+(95% CI)

Morris (1990)

GB

9.5

age

1

9 376

0.88 (0.66–1.17)

0.78 (0.54–1.12)

0.34 (0.18–0.66)

Eaton (1995)

ISR

21

age

3

8 463

0.84 (0.74–0.94)

0.81 (0.73–0.90)

0.84 (0.72–0.98)

Leon (1997)

USA

16

age

1

12 138

0.78 (0.67–0.91)

0.77 (0.66–0.90)

0.74 (0.63–0.86)

Wannamethee (1998)

GB

4

age

3

5 934

0.57 (0.40–0.78)

0.39 (0.25–0.62)

0.48 (0.34–0.68)

Morris (1990)

GB

9.5

multivariate

1

4 824

0.90 (0.57–1.44)

0.59 (0.34–1.05)

0.53 (0.21–1.32)

Sandvik (1993)

N

16

multivariate

1

1 960

0.92 (0.66–1.28)

1.00 (0.71–1.41)

0.54 (0.32–0.89)

Haapanen (1996)

FIN

11

multivariate

1

1 072

0.40 (0.21–0.75)

0.64 (0.34–1.20)

0.37 (0.19–0.68)

Folsom (1997)

USA

7

multivariate

1

6 188

0.83 (0.60–1.14)

0.97 (0.65–1.45)

0.68 (0.47–0.99)

Leon (1997)

USA

16

multivariate

1

12 138

0.85 (0.73–0.99)

0.87 (0.75–1.02)

0.83 (0.71–0.97

Erikssen (1998)

N

13

multivariate

1

1 428

0.72 (0.52–0.99)

0.48 (0.33–0.71)

0.45 (0.29–0.69)

Villeneuve (1998)

CDN

7

multivariate

1

6 246

0.81 (0.59–1.11)

0.79 (0.54–1.13)

0.86 (0.61–1.22)

Wannamethee (1998)

GB

4

multivariate

3

5 934

0.61 (0.43–0.86)

0.50 (0.31–0.79)

0.65 (0.45–0.94)

Breckenkamp (2004)

D

14

multivariate

1

2 320

0.62 (0.35–1.08)

0.76 (0.45–1.29)

0.44 (0.30–0.65)

Table A7 Studies with four intensity groups of physical activity: Sex=women.

First author (year)

Country

Follow-up (yrs)

Adjustment

Age group

Number of participants

RR in the light active group+(95% CI)

RR in the moderate active group+(95% CI)

RR in the most active group+(95% CI)

Weller (1998)

CDN

7

age

3

6 620

0.91 (0.66–1.25)

0.94 (0.72–1.23)

0.89 (0.67–1.17)

Folsom (1997)

USA

7

multivariate

1

7 852

0.79 (0.53–1.18)

1.05 (0.66–1.66)

0.58 (0.36–0.92)

Villeneuve (1998)

CDN

7

multivariate

1

8 196

0.94 (0.69–1.30)

0.92 (0.64–1.34)

0.71 (0.45–1.11)

Breckenkamp (2004)

D

14

multivariate

1

2 320

0.34 (0.14–0.81)

0.39 (0.17–0.91)

0.23 (0.12–0.42)

Table A8 Studies with five intensity groups of physical activity.

First author (year)

Country

Follow-up (year)

Sex

Adjustment

Age group

Fried (1998)

USA

5

all

age

2

Fried (1998)

USA

5

all

multivariate

2

Sundquist (2004)

S

12

all

multivariate

2

Sundquist (2004)

S

12

men

age

2

Lee (1995)

USA

26

men

multivariate

1

Sundquist (2004)

S

12

women

age

2

Rockhill (2001)

USA

20

women

age

1

Rockhill (2001)

USA

20

women

multivariate

1

First author (year)

Number of participants

RR in the light active group+(95% CI)

RR in the moderate active group+(95% CI)

RR in the active group+(95% CI)

RR in the most active group+(95% CI)

Fried (1998)

5 201

0.62

0.55

0.43

0.29

(0.49–0.80)

(0.43–0.70)

(0.34–0.55)

(0.23–0.38)

Fried (1998)

5 201

0.78

0.81

0.72

0.56

(0.60–1.00)

(0.63–1.05)

(0.55–0.93)

(0.43–0.74)

Sundquist (2004)

3 206

0.72

0.60

0.50

0.60

0.64–0.81)

(0.50–0.71)

(0.42–0.59)

(0.46–0.79)

Sundquist (2004)

1 414

0.74

0.57

0.51

0.60

(0.62–0.87)

(0.44–0.73)

(0.41–0.64)

(0.44–0.82)

Lee (1995)

17 321

0.88

0.92

0.87

0.87

(0.82–0.96)

(0.82–1.02)

(0.77–0.99)

(0.78–0.97)

Sundquist (2004)

1 792

0.70

0.59

0.47

0.54

(0.59–0.82)

(0.46–0.77)

(0.35–0.62)

(0.31–0.94)

Rockhill (2001)

4 746

0.76

0.66

0.64

0.62

(0.70–0.82)

(0.61–0.71)

(0.58–0.70)

(0.54–0.72)

Rockhill (2001)

4 746

0.82

0.75

0.74

0.71

(0.76–0.89)

(0.69–0.81)

(0.68–0.81)

(0.61–0.82)

(Above: Authors,Country,Follow-up,Sex,Adjustment, Age; Below:No of participants,RR according to the groups compared to sedentary group)

Correspondence

Prof. H. Löllgen

Department of Medicine

Ruhr-University

Bermesgasse 32 b

42897 Remscheid

Germany

Phone: +49/21916 53 54

Fax: +49/21916 106 71

Email: [email protected]