Skip to main content

Table 2 The association between physical activity and healthcare utilization, stratified by age

From: Physical activity and healthcare utilization in France: evidence from the European Health Interview Survey (EHIS) 2014

 

Total

< 65 Years Age Group

65 Years Age Group

Probit Model

Recursive Multivariate Probit Model

Probit Model

Recursive Multivariate Probit Model

Probit Model

Recursive Multivariate Probit Model

Non-prescribed medicines

 Moderately active

0.1049

p < .001

0.0455

p < .1

0.0551

.

0.1146

p < .001

0.0804

p < .1

0.0676

p < .1

 Highly active

0.1540

p < .001

0.0699

p < .1

0.1137

p < .001

0.1528

p < .001

0.1853

p < .01

0.1567

p < .1

Test of endogeneity

 

8.22

p < .001

 

7.89

p < .001

 

0.29

p < .1

Test of overidentification

 

2.85

p < .1

 

2.26

p < .1

 

1.88

p < .1

Prescribed medicines

 Moderately active

−0.0658

p < .05

− 0.0761

p < .01

− 0.0261

.

−0.0284

p < .1

−0.1349

p < .1

−0.1881

p < .1

 Highly active

−0.1064

p < .001

−0.1160

p < .001

−0.1006

p < .001

−0.1439

p < .001

−0.1273

p < .1

−0.1650

p < .1

Test of endogeneity

 

0.08

p < .1

 

2.45

p < .1

 

0.37

p < .1

Test of overidentification

 

1.01

p < .1

 

2.13

p < .1

 

1.77

p < .1

Overnight hospitalization

 Moderately active

−0.1095

p < .01

−0.1278

p < .01

−0.1397

p < .01

−0.1521

p < .001

−0.0849

p < .1

−0.0328

p < .1

 Highly active

−0.1296

p < .001

−0.1335

p < .001

−0.1048

p < .001

−0.1658

p < .001

−0.1124

p < .001

−0.0747

p < .1

Test of endogeneity

 

0.10

p < .1

 

5.31

p < .01

 

0.47

p < .1

Test of overidentification

 

1.75

p < .1

 

0.62

p < .1

   

Day hospitalization

 Moderately active

−0.0149

p < .1

− 0.0178

p < .1

− 0.0082

p < .1

−0.0811

p < .01

0.0387

p < .1

0.0811

p < .1

 Highly active

−0.0714

p < .1

−0.0316

p < .1

−0.0734

p < .1

−0.2783

p < .001

0.0484

p < .1

0.0643

p < .1

Test of endogeneity

 

0.60

p < .1

 

4.52

p < .01

 

0.19

p < .1

Test of overidentification

 

1.70

p < .1

 

1.72

p < .1

 

0.40

p < .1

Generalist physician services

 Moderately active

0.2090

p < .001

0.2346

p < .001

0.2264

p < .001

0.1802

p < .001

0.0178

p < .1

0.2615

p < .01

 Highly active

0.2106

p < .001

0.2937

p < .001

0.2393

p < .001

0.1917

p < .001

0.0200

p < .1

0.2613

p < .01

Test of endogeneity

 

8.18

p < .001

 

5.11

p < .01

 

6.16

p < .01

Test of overidentification

   

 

1.32

p < .1

 

0.66

p < .1

Specialist physician services

 Moderately active

−0.0343

p < .1

−0.0310

p < .1

0.0430

p < .1

0.0152

p < .1

0.0019

p < .1

0.0397

p < .1

 Highly active

−0.0415

p < .1

−0.1286

p < .001

0.0482

p < .1

0.0118

p < .1

0.0385

p < .1

0.1309

p < .1

Test of endogeneity

 

0.93

p < .1

 

2.42

p < .1

 

0.65

p < .1

Test of overidentification

 

2.16

p < .1

 

0.99

p < .1

 

0.66

p < .1

Preventive services

 Moderately active

0.0719

p < .01

0.0501

p < .1

0.0851

p < .01

0.0660

p < .05

−0.0343

p < .1

−0.0331

p < .1

 Highly active

0.0826

p < .01

0.0657

p < .1

0.0880

p < .01

0.1185

p < .01

−0.0683

p < .1

−0.0574

p < .1

Test of endogeneity

 

0.13

p < .1

 

0.16

p < .1

 

0.03

p < .1

Test of overidentification

 

3.26

.

 

0.45

p < .1

 

3.11

p < .1

Home healthcare services

 Moderately active

−0.2152

p < .001

−0.2298

p < .001

−0.1813

p < .001

−0.1921

p < .001

−0.2137

p < .001

−0.2599

p < .001

 Highly active

−0.2599

p < .001

−0.3151

p < .001

−0.2044

p < .001

−0.2247

p < .001

−0.3179

p < .001

−0.4213

p < .001

Test of endogeneity

 

26.09

p < .001

 

16.46

p < .001

 

2.73

p < .05

Test of overidentification

 

3.08

p < .1

 

2.32

p < .1

 

0.46

p < .1

  1. (1) The table reports estimates for two approaches (probit and recursive multivariate probit) on the association between physical activity and the use of medicines and healthcare services. The coefficients indicate the difference between the probit index (Z-score) of each category of physical activity, i.e. moderately and highly physically active, and that corresponding to the reference group for physical activity, which is low physically active. (2) The relationship between physical activity and healthcare utilization is controlled by the following variables: sex of respondent; age group; education level; legal marital status; employment status; insurance; income level; BMI status; smoking; alcohol consumption risk profile; nutrition – fruits and vegetables consumption; health status index
  2. Source: Authors’ computation