X hits on this document

58 views

0 shares

0 downloads

0 comments

17 / 22

CVD

-0.106

0.060

-0.088

0.449

Diabetes

-0.058

0.319

-0.124

0.003

High school

0.047

0.071

0.119

0.089

University

0.059

0.010

0.211

0.054

Age

0.257

0.000

0.990

0.070

Age2

-0.039

0.000

-0.124

0.031

Married

0.039

0.008

-0.046

0.336

Children

0.033

0.114

-0.163

0.055

Treated hypertension

-0.002

0.038

0.007

0.859

Lipid therapy

-0.007

0.010

0.018

0.860

Sufficient exercise

0.002

0.075

-0.002

0.958

Obesity

-0.010

0.000

-0.023

0.749

Current smoking

-0.006

0.102

0.008

0.362

Parent diabetic

-0.006

0.003

-0.015

0.763

Predicted prob. of participation

0.869

0.012

0.483

0.000

Males

Females

n=4272

n=5251

-0.223

0.021

-0.476

0.620

Correlation. Coefficient

p

Correlation. Coefficient

p

0.080 0.364 0.117

0.643 0.050 0.463 0.000

-0.281 0.157 -0.647

0.366 0.359 0.000 0.000

Table 5: Marginal effect of risk factors on labour force participation in males and females aged 25 and over: results from multivariate regressions

Multivariate probit

Prob.(LFP=1|CVD=1,Diab=1) - P(LFP=1|CVD=0,Diab=0)

Variables

1,2

p

1,3

Marginal

1=Participation; 2= Cardiovascular disease(CVD); 3= Diabetes; p value for t-test on marginal effects and Chi2 for correlations

p

Marginal effect

1,2=

2,3

1,3=

2,3=0

Table 5 also shows the marginal indirect effect of the disease risk factors on labour force participation. For example the effect on males of being obese is to increase the risk of diabetes and cardiovascular disease and thereby indirectly reduce the probability of being in the labour force by 0.01. Similarly smoking, insufficient exercise, hypertension, and abnormal cholesterol each have a small but significant effect on labour force participation for both men and women through their effect on the risk of chronic disease.

In addition to the effect on labour force participation it is likely that diabetes and cardiovascular disease have an effect on the labour productivity of those in employment. The AusDiab survey does not have information on wages or hours worked to allow us to test this. It is also likely that the severity of disease has an impact on labour force participation. Including self reported general morbidity in a system of equations would considerably increase the complexity of the estimation in the multivariate system. Instead we explored this issue by including a dichotomous variable for self

Chronic disease and labour force participation in Australia: an endogenous multivariate probit analysis of clinical prevalence data

13

Document info
Document views58
Page views58
Page last viewedThu Dec 08 06:57:19 UTC 2016
Pages22
Paragraphs729
Words8718

Comments