4727
Views
1010
Downloads |
Why are older informal carers in better health? Solving a causality problem
Preliminary communication | Year: 2018 | Pages: 303 - 323 | Volume: 42 | Issue: 3 Received: May 23, 2018 | Accepted: July 12, 2018 | Published online: September 3, 2018
|
FULL ARTICLE
FIGURES & DATA
REFERENCES
CROSSMARK POLICY
METRICS
LICENCING
PDF
|
|
Country
|
Helpgiving_tot
|
Helpgiving_out
|
Helpgiving_wtin
|
|
Scand
|
Sweden
|
1,527,567
|
1,460,927
|
113,159
|
|
Denmark
|
1,048,569
|
1,005,553
|
92,927
|
|
Netherlands
|
2,376,857
|
2,163,370
|
315,709
|
|
Contin
|
Austria
|
953,164
|
833,773
|
177,566
|
|
Belgium
|
1,637,570
|
1,426,540
|
331,931
|
|
France
|
8,371,773
|
7,252,966
|
1,507,020
|
|
Germany
|
12,648,003
|
11,455,470
|
1,745,328
|
|
Luxembourg
|
52,710
|
45,376
|
10,395
|
|
Switzerland
|
839,445
|
790,900
|
90,097
|
|
Medit
|
Italy
|
7,153,458
|
6,053,197
|
1,691,223
|
|
Spain
|
3,572,164
|
2,512,323
|
1,268,977
|
|
East
|
Czech Republic
|
1,483,696
|
1,341,115
|
298,467
|
|
Estonia
|
172,302
|
150,356
|
36,329
|
|
Slovenia
|
131,627
|
106,774
|
39,095
|
|
Mix
|
Israel
|
306,143
|
198,250
|
119,050
|
|
Tot
|
Total
|
42,275,048
|
36,796,889
|
7,837,273
|
|
Probit: Help_outside
|
Coeff.
|
z
|
P>z
|
Coeff.
|
z
|
P>z
|
Coeff.
|
z
|
P>z
|
|
Gender
|
0.0095
|
0.84
|
|
0.0104
|
0.91
|
|
-0.0016
|
-0.14
|
|
|
Age
|
-0.0239
|
-28.51
|
***
|
-0.0234
|
-28.12
|
***
|
-0.0233
|
-27.80
|
***
|
|
Edu_Years
|
0.0190
|
13.59
|
***
|
0.0190
|
13.61
|
***
|
0.0191
|
13.63
|
***
|
|
Income_Middle
|
0.0371
|
2.67
|
***
|
0.0373
|
2.69
|
***
|
0.0392
|
2.81
|
***
|
|
Income_Upper
|
0.1028
|
7.19
|
***
|
0.1026
|
7.16
|
***
|
0.1077
|
7.48
|
***
|
|
Retired vs. Employed
|
0.0789
|
4.76
|
***
|
0.0818
|
4.94
|
***
|
0.0789
|
4.75
|
***
|
|
Other vs. Employed
|
-0.0225
|
-1.20
|
|
-0.0162
|
-0.87
|
|
-0.0260
|
-1.38
|
|
|
Hh_Size
|
-0.0514
|
-8.12
|
***
|
-0.0513
|
-8.11
|
***
|
-0.0510
|
-8.02
|
***
|
|
Physical_Inactivity
|
-0.3691
|
-16.73
|
***
|
-0.3638
|
-16.52
|
***
|
-0.3883
|
-17.27
|
***
|
|
Memory
|
0.0377
|
13.23
|
***
|
0.0369
|
12.91
|
***
|
0.0384
|
13.33
|
***
|
|
Continental
|
0.1064
|
7.68
|
***
|
0.1045
|
7.51
|
***
|
0.1037
|
7.43
|
***
|
|
Socialdemocratic
|
0.4534
|
25.15
|
***
|
0.4553
|
24.65
|
***
|
0.4624
|
25.46
|
***
|
|
Mediterranean
|
-0.1866
|
-9.79
|
***
|
-0.1893
|
-9.93
|
***
|
-0.1897
|
-9.90
|
***
|
|
Chronic diseases
|
0.0404
|
3.50
|
***
|
|
|
|
|
|
|
|
Self-rated Health
|
|
|
|
0.0077
|
0.60
|
|
|
|
|
|
Depression
|
|
|
|
|
|
|
0.1052
|
7.93
|
***
|
|
Constant
|
0.5855
|
8.89
|
***
|
0.5676
|
8.53
|
***
|
0.5541
|
8.35
|
***
|
|
Observations
|
62257
|
|
|
62330
|
|
|
61547
|
|
|
|
LR Chi2
|
5073.81
|
***
|
|
5032.93
|
***
|
|
5001.36
|
***
|
|
|
Pseudo R2
|
0.0673
|
|
|
0.0667
|
|
|
0.0670
|
|
|
|
Log Likelihood
|
-35168.70
|
|
|
-35221.55
|
|
|
-34847.12
|
|
|
|
Help_total
|
Coeff.
|
z
|
P>z
|
Coeff.
|
z
|
P>z
|
Coeff.
|
z
|
P>z
|
|
Gender
|
0.0246
|
2.72
|
***
|
0.0183
|
2.11
|
**
|
0.0282
|
2.82
|
***
|
|
Age
|
-0.0071
|
-10.03
|
***
|
-0.0078
|
-12.34
|
***
|
-0.0079
|
-12.62
|
***
|
|
Edu_Years
|
0.0030
|
2.86
|
***
|
0.0028
|
2.64
|
***
|
0.0032
|
3.07
|
***
|
|
Income_Middle
|
0.0133
|
1.33
|
|
0.0108
|
1.09
|
|
0.0132
|
1.33
|
|
|
Income_Upper
|
0.0161
|
1.45
|
|
0.0124
|
1.10
|
|
0.0174
|
1.57
|
|
|
Retired vs. Employed
|
0.0258
|
1.69
|
*
|
0.0159
|
1.16
|
|
0.0068
|
0.52
|
|
|
Other vs. Employed
|
0.0112
|
0.60
|
|
-0.0013
|
-0.08
|
|
-0.0088
|
-0.56
|
|
|
Hh_Size
|
0.0026
|
0.48
|
|
0.0022
|
0.40
|
|
0.0021
|
0.37
|
|
|
Physical_Inactivity
|
-0.0391
|
-2.30
|
**
|
-0.0504
|
-3.29
|
***
|
-0.0439
|
-2.62
|
***
|
|
Memory
|
0.0061
|
2.69
|
***
|
0.0060
|
2.60
|
***
|
0.0065
|
2.86
|
***
|
|
Continental
|
-0.0140
|
-0.77
|
|
-0.0042
|
-0.25
|
|
0.0037
|
0.22
|
|
|
Socialdemocratic
|
0.0652
|
3.42
|
***
|
0.0512
|
2.40
|
**
|
0.0784
|
4.36
|
***
|
|
Mediterranean
|
-0.0669
|
-3.41
|
***
|
-0.0553
|
-3.00
|
***
|
-0.0471
|
-2.53
|
**
|
|
Chronic diseases
|
-0.1309
|
-2.49
|
**
|
|
|
|
|
|
|
|
Self-rated Health
|
|
|
|
-0.0936
|
-2.46
|
**
|
|
|
|
|
Depression
|
|
|
|
|
|
|
-0.0739
|
-1.97
|
**
|
|
Constant
|
0.7908
|
14.58
|
***
|
0.8617
|
14.07
|
***
|
0.7942
|
14.63
|
***
|
|
Observations
|
13232
|
|
|
13179
|
|
|
13149
|
|
|
|
Wald Chi2
|
650.65
|
***
|
|
660.09
|
***
|
|
654.60
|
***
|
|
|
Log Likelihood
|
-17116.06
|
|
|
-15074.79
|
|
|
-14999.20
|
|
|
|
Test of endogeneity
|
7.04
|
***
|
|
4.87
|
**
|
|
8.52
|
***
|
|
|
Help_outside
|
Coeff.
|
z
|
P>z
|
Coeff.
|
z
|
P>z
|
Coeff.
|
z
|
P>z
|
|
Gender
|
0.0136
|
1.57
|
|
0.0055
|
0.66
|
|
0.0235
|
2.5
|
**
|
|
Age
|
-0.0083
|
-12.56
|
***
|
-0.0093
|
-15.51
|
***
|
-0.0095
|
-15.75
|
***
|
|
Edu_Years
|
0.0039
|
3.92
|
***
|
0.0038
|
3.81
|
***
|
0.0041
|
4.14
|
***
|
|
Income_Middle
|
0.0096
|
1.00
|
|
0.0070
|
0.74
|
|
0.0077
|
0.8
|
|
|
Income_Upper
|
0.0163
|
1.53
|
|
0.0144
|
1.35
|
|
0.0165
|
1.55
|
|
|
Retired vs. Employed
|
0.0333
|
2.34
|
**
|
0.0156
|
1.21
|
|
0.0086
|
0.68
|
|
|
Other vs. Employed
|
0.0131
|
0.76
|
|
-0.0092
|
-0.59
|
|
-0.0103
|
-0.68
|
|
|
Hh_Size
|
-0.0275
|
-5.19
|
***
|
-0.0277
|
-5.23
|
***
|
-0.0288
|
-5.39
|
***
|
|
Physical_Inactivity
|
-0.0488
|
-3.07
|
***
|
-0.0690
|
-4.78
|
***
|
-0.0465
|
-2.94
|
***
|
|
Memory
|
0.0086
|
3.97
|
***
|
0.0092
|
4.28
|
***
|
0.0082
|
3.77
|
***
|
|
Continental
|
-0.0139
|
-0.81
|
|
0.0033
|
0.21
|
|
0.0131
|
0.83
|
|
|
Socialdemocratic
|
0.0780
|
4.31
|
***
|
0.0727
|
3.74
|
***
|
0.0965
|
5.6
|
***
|
|
Mediterranean
|
-0.0799
|
-4.30
|
***
|
-0.0624
|
-3.56
|
***
|
-0.0491
|
-2.75
|
***
|
|
Chronic diseases
|
-0.1749
|
-3.92
|
***
|
|
|
|
|
|
|
|
Self-rated Health
|
|
|
|
-0.0839
|
-2.74
|
**
|
|
|
|
|
Depression
|
|
|
|
|
|
|
-0.1404
|
-4.24
|
***
|
|
Constant
|
0.9109
|
17.49
|
***
|
0.9716
|
17.21
|
***
|
0.9227
|
17.71
|
***
|
|
Observations
|
13236
|
|
|
13183
|
|
|
13153
|
|
|
|
Wald Chi2
|
1036.86
|
***
|
|
1062.89
|
***
|
|
1053.86
|
***
|
|
|
Log Likelihood
|
-16505.67
|
|
|
-14463.15
|
|
|
-14398.63
|
|
|
|
Test of endogeneity
|
14.92
|
***
|
|
5.08
|
**
|
|
18.90
|
***
|
|
|
Help_within
|
Coeff.
|
z
|
P>z
|
Coeff.
|
z
|
P>z
|
Coeff.
|
z
|
P>z
|
|
Gender
|
0.0297
|
2.63
|
***
|
0.0136
|
1.34
|
|
0.0281
|
2.76
|
***
|
|
Age
|
0.0018
|
2.36
|
**
|
0.0013
|
1.86
|
*
|
0.0019
|
2.91
|
***
|
|
Edu_Years
|
-0.0010
|
-0.74
|
|
-0.0014
|
-1.12
|
|
-0.0004
|
-0.33
|
|
|
Income_Middle
|
0.0042
|
0.34
|
|
-0.0083
|
-0.73
|
|
0.0002
|
0.02
|
|
|
Income_Upper
|
-0.0237
|
-1.65
|
*
|
-0.0407
|
-3.15
|
***
|
-0.0131
|
-1.09
|
|
|
Retired vs. Employed
|
0.0682
|
3.56
|
***
|
0.0382
|
2.23
|
**
|
-0.0008
|
-0.05
|
|
|
Other vs. Employed
|
0.0833
|
3.96
|
***
|
0.0509
|
2.70
|
***
|
0.0030
|
0.18
|
|
|
Hh_Size
|
0.0182
|
3.01
|
***
|
0.0188
|
3.25
|
***
|
0.0273
|
4.19
|
***
|
|
Physical_Inactivity
|
0.0518
|
3.32
|
***
|
0.0389
|
2.77
|
***
|
0.0086
|
0.62
|
|
|
Memory
|
-0.0055
|
-1.97
|
**
|
-0.0055
|
-2.21
|
**
|
-0.0011
|
-0.46
|
|
|
Receiving_help
|
0.1061
|
7.36
|
***
|
0.0819
|
5.88
|
***
|
0.0902
|
5.87
|
***
|
|
Continental
|
-0.1043
|
-5.39
|
***
|
-0.0729
|
-4.20
|
***
|
-0.0532
|
-3.32
|
***
|
|
Socialdemocratic
|
-0.1090
|
-4.92
|
***
|
-0.1615
|
-8.06
|
***
|
-0.0792
|
-4.32
|
***
|
|
Mediterranean
|
-0.0548
|
-2.54
|
**
|
-0.0404
|
-2.09
|
**
|
-0.0260
|
-1.41
|
|
|
Chronic diseases
|
-0.4332
|
-34.91
|
***
|
|
|
|
|
|
|
|
Self-rated Health
|
|
|
|
-0.4326
|
-36.13
|
***
|
|
|
|
|
Depression
|
|
|
|
|
|
|
0.0119
|
0.36
|
|
|
Constant
|
0.2021
|
3.08
|
***
|
0.3982
|
6.60
|
***
|
-0.0962
|
-1.68
|
*
|
|
Observations
|
4656
|
|
|
4634
|
|
|
4615
|
|
|
|
Wald Chi2
|
1318.33
|
***
|
|
1397.63
|
***
|
|
136.85
|
***
|
|
|
Log Likelihood
|
-3592.05
|
|
|
-2002.05
|
|
|
-3583.52
|
|
|
|
Test of endogeneity
|
299.29
|
***
|
|
525.46
|
***
|
|
0.38
|
|
|
Source: Own calculations.
|
|
Help_total
|
Help_outside
|
Help_within
|
|
Chronic diseases
|
-0.1545
|
***
|
-0.1837
|
***
|
-0.5459
|
***
|
|
Self-rated Health
|
-0.1095
|
***
|
-0.1108
|
***
|
-0.5537
|
***
|
|
Depression
|
-0.0939
|
**
|
-0.1559
|
***
|
0.0130
|
|
|
|
|
|
|
|
|
|
|
Help_total
|
Help_outside
|
Help_within
|
|
Chronic diseases
|
-0.1493
|
***
|
-0.1924
|
***
|
-0.6022
|
***
|
|
Self-rated Health
|
-0.1264
|
***
|
-0.0915
|
**
|
-0.5451
|
***
|
|
Depression
|
-0.0806
|
*
|
-0.1685
|
***
|
-0.0687
|
*
|
|
|
|
|
|
|
|
|
|
Help_total
|
Help_outside
|
Help_within
|
|
Chronic diseases
|
-0.0904
|
*
|
-0.0910
|
*
|
-0.1170
|
*
|
|
Self-rated Health
|
-0.0243
|
|
-0.0352
|
*
|
-0.3504
|
**
|
|
Depression
|
-0.0569
|
*
|
-0.0857
|
*
|
0.0025
|
|
|
|
Effects of health
variables
|
Effects of confounders
|
Reverse causality
issues
|
|
Help-giving Total
|
All coefficients on health variables,
which are of our main interest, are of the expected, negative sign and
significant; but only after controlling for reverse causality
|
Significant and of expected sign:
Gender, age, education, physical inactivity, memory, welfare regimes
|
Only with the inclusion of the
instruments for each health variable, are
the results as expected
|
|
Help-giving Outside
|
All coefficients on health variables,
which are of our main interest, are of the expected, negative sign and
significant; but only after controlling for reverse causality
|
Significant and of expected sign: Age,
education, household size, physical inactivity, memory, welfare regimes
|
Only with the inclusion of the
instruments for each health variable, are
the results as expected
|
|
Help-giving Within
|
All coefficients on health variables,
which are of our main interest, are of the expected, negative sign and
significant; but only after controlling for reverse causality and including
an auxiliary variable
|
Significant and of expected sign:
Gender, age, income, employment status, household size, physical inactivity,
memory, welfare regimes
|
One does not observe the expected
relationships even after endogeneity is provided for by the instrumental
variable correction. Only with the final intervention, inclusion of the
auxiliary variable of receiving is care within household the issue solved,
providing the (negative) sign and significance of almost all of the
coefficients on health variables.
|
Graph 1Distribution of the help-giving variables (%) DISPLAY Graph
Table 1Estimation of total help-givers, Deville-Särndal’s procedure DISPLAY Table
Graph 2Distribution of the main three health variables (left: number of chronic diseases; middle: self-rated health; right: EURO-Depression scale), based on help-giving within and outside the household and total help-giving DISPLAY Graph
Table 2Results of the models with no endogeneity provided for DISPLAY Table
Graph 3Modelling of reverse causality in the model DISPLAY Graph
Table 3Results of models, endogeneity controlled for, total help-giving DISPLAY Table
Table 4Results of models, endogeneity controlled for, help-giving outside a household DISPLAY Table
Table 5Results of models, endogeneity controlled for, help-giving within household DISPLAY Table
Table 6 Results of robustness tests, top: exclusion of additional health variables due to additional endogeneity problems; middle: restricting the age of the respondents: 65+, bottom: including an additional instrument to control for reverse causality between providing and receiving help within a household DISPLAY Table
Table 6 middle: restricting the age of the respondents: 65+, bottom: including an additional instrument to control for reverse causality between providing and receiving help within a household DISPLAY Table
Table 7Summarized main results of the paper by type of help-giving DISPLAY Table
* We would like to thank the two anonymous referees for valuable comments and suggestions.
1 All of those hypotheses relate to decision of providing care and not to quality or type of care.
2 In the article, we use the terms caregiving and help-giving as synonymous.
3 The studies also report a noticeable difference in respondents’ answers about the type of help received from other people (e.g. personal care vs. practical household help, see Hoefman, Meulenkamp and de Jong,  2017). In providing care within the household, assistance is related to personal care. Help-giving outside a household refers to personal care and practical care.
4 This paper uses data from SHARE Wave 5, see Börsch-Supan et al. ( 2013) for methodological details. The SHARE data collection has been primarily funded by the European Commission through FP5 (QLK6-CT-2001-00360), FP6 (SHARE-I3: RII-CT-2006-062193, COMPARE: CIT5-CT-2005-028857, SHARELIFE: CIT4-CT-2006-028812) and FP7 (SHARE-PREP: N°211909, SHARELEAP: N°227822, SHARE M4: N°261982). Additional funding from the German Ministry of Education and Research, the U.S. National Institute on Aging (U01_AG09740-13S2, P01_AG005842, P01_AG08291, P30_AG12815, R21_AG025169, Y1-AG-4553-01, IAG_BSR06-11, OGHA_0
5 For more details, see Börsch-Supan et al., 2013; 2015; Malter and Börsch-Supan, 2015; Börsch-Supan, 2016.
6 For the explanation of the two variables, see descriptions above.
7 In SHARE, there is also the variable of delayed recall of words which is not used here – but robustness checks have been done using this variable as a control as well with no significant changes in the results.
8 An additional check for the group of 80+ has been done with no changes in results.
|
|
September, 2018
III/2018
|
