Chapter 4 Relationship between SES, Mental Health and Need for LTC among the
4.3 Results
4.3.2 NLTC of the elderly in Yanji
114
year, but negatively correlated with the duration on bed. A significant positive correlation was observed between income and current health status, health status compared with elderly surrounded; moreover, a significant negative correlation was also found between income and duration on bed, extent should be taken care. Three variables (current health status, health status compared with elderly surrounded and health status compared with last year) were found to be negatively correlated with duration on bed and extent should be taken care in pairs.
115
As for the second care provider, spouse is the most important one since it constituted for 71.9 percent, followed by the son with the percentage of 22.5 percent.
While for the expected second care provider, spouse was the most desired one (with a proportion of 81.4%), follow by the son (16.9%).
Concerning the third care provider, son played the most important role and the proportion was 71.4%, follow by daughter (19.0%). Son was also expected to be the most important third care providers, since its percentage rank highest among the expected third care provider and accounted for 67.5 percent, while the role of other care providers as the third provider (such as the daughter and daughter-in-law) is relatively lower.
4.3.3 Structural relationship between SES, mental health and NLTC among the elderly in Yanji city
4.3.3.1 Factor analysis
Table 4.5 shows the results of the factor analysis. After the principal component analysis by using the rotation method of varimax with kaiser normalization, seven variables (“Current health status”, “Health status compared with last year”, “Health status compared with elderly surrounded”, “Extent should be taken care”, “Duration on bed”, “Income” and “Education”) were divided into three main factors with a cumulative contribution rate of 67.370%.
Table 4.5 Factor Analysis of Observed Variables Component
Mental Health Need for LTC Socioeconomic status
Current Health Status .840 -.067 .103
Health Status Compared With Last Year .788 -.094 -.037
Health Status Compared With Elderly Surrounded .617 -.086 .353
Extent Should Be Taken Care -.019 .850 .089
Duration On Bed -.354 .640 -.037
Income -.118 -.624 .618
Education .222 .113 .859
Cumulative % 27.089 49.301 67.370
Cronbach's Alpha Reliability Statistics .682 .676 .629
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization.
a. Rotation converged in 5 iterations.
The three main factors were named as “Mental Health” (including three variables:
Current health status, Health status compared with last year, Health status compared
116
with elderly surrounded), “Need for LTC” (including two variables: Extent should be taken care, Duration on bed) and “Socioeconomic status” (including two variables:
Education, Income).
The coefficient for the Cronbach's Alpha Reliability Statistics of the “Mental Health”, “Need for LTC” and “SES” was 0.682, 0.676 and 0.629 respectively. All of them are higher than 0.6, which indicate the internal consistency of these three main factors was relatively acceptable.
4.3.3.2 Introduction of the SEM
Structural equation modeling (referred to as SEM) is a useful method to help the researchers to explore the complex causal relationships between multiple variables, especially, for those cannot be directly observed and the indirect relationships. Two forms of variables are contained in SEM, named latent variable and observed variable separately. To present them clearly, ellipse is used to represent the latent variables and rectangle for the observed variables. When conducting the SEM, the first step is to extract the common factors from the related observed variables by performing the factor analysis. Then, regression analysis is needed to conduct on these extracted the common factors (used as the latent variables), with the purpose of investigating the structural relationships between them. In the SEM, the arrow is used to represent the relationship between two variables, and the value on the arrows is the standardized path coefficient (between -1 to 1) which indicates the magnitude and direction of the relationship. Moreover, the variation of the observed variables that cannot be explained by the latent variable is represented by “e”, while the variation of the latent variable that cannot be explained by other latent variable is represented by “z” in our study.
In this study, “Socioeconomic status”, “Mental health” and “Need for LTC”, these three variables are the latent variables; while “Current health status”, “Health status compared with last year”, “Health status compared with elderly surrounded”, “Extent should be taken care”, “Duration on Bed”, “Income” and “Education”, these seven variables were the observed variables. By employing the SEM, we aim to find out both the direct and indirect relationships between all of these variables.
4.3.3.3 Fitness of the SEM
The model fitness indices were shown in Table 4.6. The values of the four fitness indexes (CFI, TLI, IFI and RMSEA) were 0.987, 0.956, 0.988 and 0.039 respectively.
They all meet the recommended criteria for the model fitness indices, suggesting that the proposed model fit the original data well. The p-value of the Chi-square test (0.216) was higher than the recommended level and also indicated the proposed model fit with the original data ideally.
117
Table 4.6 Model fitness indices for the hypothetical model of the elderly in Yanji
Index p CFI TLI IFI RMSEA
Range - 0-1 0-1 0-1 -
Criteria(Recommended Level) P>0.05 ≥0.9 ≥0.9 ≥0.9 ≤0.05
Hypothetical(Calculated value) 0.216 0.987 0.956 0.988 0.039
4.3.3.4 Relationship between mental health and NLTC
A negative effect of mental health on the need for LTC was observed (Table 4.7 and Figure 4.1), with a path coefficient of 0.278 ≈ -0.28. Current health status had the greatest effect on the need for LTC, with a path coefficient of 0.73*-0.28. Current health status compared with last year and the current health status compared with elderly surrounded had a smaller effect on the need for LTC than the current health status, with a path coefficient equals of 0.67*-0.28 and 0.55*-0.28, respectively.
These results indicated a negative relationship between mental health and the need for LTC. People with good mental health will have a low need for LTC. The opposite will also be true: poor health individuals will have a high need for LTC. Thus, the hypothesis 2 has been verified.
4.3.3.5 Correlation between SES and NLTC
Socioeconomic status could influence the need for LTC in two ways: directly and indirectly via mental health. As seen in Table 4.7 and Figure 4.1, the direct impact of socioeconomic status on the need for LTC was relatively weak (path coefficient equals -0.21), while the indirect effect of socioeconomic status on the need for LTC via mental health was also weak (path coefficient equals 0.45 * -0.28 ≈ -0.13). It is worth noting that although both the direct and indirect effect was weak, they were both significant. In summary, socioeconomic status had both direct and indirect inverse effect on the need for LTC.
4.3.3.6 Association between SES and mental health
As demonstrated in Table 4.8 and Figure 4.1, the path coefficient between socioeconomic status and mental health is 0.448 ≈ 0.45. It indicates that socioeconomic status of the elderly in Yanji had a moderate positive effect on mental
Table 4.7 Standardized effects of SES and Mental Health on NLTC of the elderly in Yanji
Variable Direct Indirect Total
SES -0.207 -0.125 -0.332
Mental Health -0.278 - -0.278
Note: Dependent Variable: NLTC; Independent Variables: SES, Mental Health; NLTC=need for long-term care; SES=
socioeconomic status.
118
health: the higher socioeconomic status, the better mental health. Thus, Hypothesis 1 has been verified. Of the indicators of SES, education had a greater impact on mental health (path coefficient equals 0.86 * 0.45); while the effect from income onto the mental health was relatively low (0.25 * 0.45).
Table 4.8 Standardized effects of SES on Mental Health of the elderly in Yanji
Variable Direct Indirect Total
Mental Health 0.448 - 0.448
Note: Dependent Variable: NLTC; Independent Variables: SES, Mental Health; NLTC=need for long-term care; SES=
socioeconomic status.
4.3.4 Gender difference of the relationship between SES, mental health and NLTC 4.3.4.1 Model fitness indices
Table 4.9 Model fitness indices for the hypothetical model of the elderly in Yanji by gender
Index p CFI TLI IFI RMSEA
Range - 0-1 0-1 0-1 -
Criteria(Recommended Level) P>0.05 ≥0.9 ≥0.9 ≥0.9 ≤0.05
Hypothetical(Calculated value) 0.052 0.940 0.888 0.946 0.045
Figure 4.1 Structural relationships between the SES, mental health and NLTC among the elderly in Yanji
.20 Mental Health
Socioeconomic Status
.06 Income
e7 .75
Education
e6 .45
z1
CMIN=10.750 P=.216 CFI=.987 TLI=.956 IFI=.988
RMSEA=.039
.17 Need for LTC
.55
Duration On Bed
e4
.16
E xtent S hould Be Taken Care
e5
z2
.53
Current Health S tatus
e1
.45
Health S tatus Com pared With Last Y ear
e2
.30
Health S tatus Com pared With E lderly S urrounded
e3
-.28
-.21
.25 .67 .74
-.34
-.40 .41
.17 .86
.55 .73
119
Figure 4.2 Structural relationships between the SES, mental health and NLTC among the male elderly in Yanji
.24 Mental Health
Socioeconomic Status
.11 Income
e7 .67
Education
e6
z1 .49
CMIN=43.624 P=.052 CFI=.940 TLI=.888 IFI=.946
RMSEA=.045 Group=Male elderly
.22 Need for LTC
.58
Duration On Bed
e4
.19
E xtent S hould Be Taken Care
e5
z2
.51
Current Health S tatus
e1
.53
Health S tatus Com pared With Last Y ear
e2
.42
Health S tatus Com pared With E lderly S urrounded
e3
-.25
-.29
.33 .73 .76
-.56
-.32 .44
.11 .82
.65 .71
Figure 4.3 Structural relationships between the SES, mental health and NLTC among the female elderly in Yanji .18
Mental Health
Socioeconomic Status
.05
Income
e7 .51
Education
e6
z1 .42
CMIN=43.624 P=.052 CFI=.940 TLI=.888 IFI=.946
RMSEA=.045 Group=Female elderly
.16 Need for LTC
.50
Duration On Bed
e4
.16
E xtent S hould Be Taken Care
e5
z2
.43
Current Health S tatus
e1
.43
Health S tatus Com pared With Last Y ear
e2
.20
Health S tatus Com pared With E lderly S urrounded
e3
-.24
-.24
.23 .66 .71
-.14
-.46 .40
.29 .72
.44 .66
120
The model fitness indices for the hypothetical model of the elderly in Yanji by gender were shown in Tables 4.9. The recommended level for the Chi-square test is P>0.05 in this study, while the calculated value was 0.052, implying the hypothetical model fit the empirical data well. The absolute value of other three fitness indexes of CFI, IFI and RMSEA were 0.940, 0.946 and 0.045 respectively. They all meet the recommended criteria of the model fitness indices, which meant that the proposed model ideally fit with the original data. The TLI (0.888) was slightly lower than the recommended level of 0.9, and was fairly acceptable to be used to indicate the good fitness of the model.
4.3.4.2 Gender difference on the association between mental health and NLTC
The gender difference of the relationship between mental health and NLTC of the elderly in Yanji was presented in Table 4.10, Figure 4.2 and Figure 4.3. A negative association was observed in both male and female elderly in Yanji. The standardized effect of mental health on the NLTC was -0.255 in the male elderly and -0.236 in the female elderly, implying the relationship between mental health and the NLTC was slightly stronger in men than in women.
Table 4.10 Standardized effects of SES and Mental Health on NLTC among the elderly in Yanji by gender
Variable
Direct Indirect Total
Men Women Men Women Men Women
SES -0.292 -0.236 -0.124 -0.100 -0.416 -0.336
Mental Health -0.255 -0.236 - - -0.255 -0.236
Note: Dependent Variable: NLTC; Independent Variables: SES, Mental Health
4.3.4.3 Gender difference on the relationship between SES and NLTC
The difference on the relationship between SES and the NLTC among the male elderly and female elderly in Yanji was shown in Table 4.10, Figure 4.2 and Figure 4.3. SES had both direct and indirect on NLTC in both men and women. In general, the standardized direct effect (-0.292 for men and -0.236 for women) was stronger than the indirect one (-0.124 for men and -0.100 for women) in both male elderly and female elderly.
Both the standardized direct and indirect relationship was stronger in men (-0.292 for the direct standardized effect and -0.124 for the indirect standardized effect) than in women (-0.236 for the direct standardized effect and -0.100 for the indirect standardized effect). Thus, the total standardized effect was stronger in men than in women (-0.416 vs -0.336).
4.3.4.3 Gender difference on the relationship between SES and mental health
121
Table 4.11 Standardized effects of SES on Mental Health among the elderly in Yanji by gender
Variable
Direct Indirect Total
Men Women Men Women Men Women
Mental Health 0.485 0.423 - - 0.485 0.423
Note: Dependent Variable: NLTC; Independent Variables: SES, Mental Health
The comparison on the standardized effects of SES on mental health among the elderly in Yanji was illustrated in Table 4.11, Figure 4.2 and Figure 4.3. A positive association was found between SES and mental health in both the male and female elderly in Yanji. Specifically, SES had a slightly stronger influence on mental health in men than in women (0.485 for the male elderly and 0.423 for the female elderly).
4.4 Discussion
4.4.1 Discussion on the satisfaction of the NLTC
The actual care provider and the expected care provider are highly in accordance.
This indicated that the need for LTC of the elderly in Yanji city was well satisfied.
Moreover, the important role of the family members as the informal care provider on providing the LTC was also demonstrated in this study.
4.4.2 Relationship between mental health and NLTC
A negative correlation between mental health and the need for LTC is the most important finding of this study, because few researches have explored the relationship between mental health and the need for LTC. Many existing studies had mainly investigated the need for LTC among the elderly who suffered from physical disability. For example, Wolfe et al. conducted a 10-year follow-up study on the need for LTC of the stroke patients in south London of UK (Wolfe et al., 2011). Among the few studies involving the relationship between mental health and the need for LTC, researchers only provided the distribution of need for LTC of the elderly, without exploring the relationship between mental health and need for LTC. For example, Kim et al. studied the need for LTC of the patients with dementia (Kim et al., 2011);
Reid et al. investigated the influencing factors of the mental disorders and the need for LTC by the adult patients (Reid et al., 2011). In addition, the study also explained why there were 53 percent of the patients with dementia, 34 percent with mental health problems and only 12 percent were in good mental health among the patients who were using the home care and institutional nursing care services in one study.
This is because there is a negative correlation between mental health and the need for LTC.
122 4.4.3 Correlation between SES and NLTC
The relationship between SES and NLTC was found to be negative. The result means that individual with a high socioeconomic status will have a low need for LTC;
those with lower socioeconomic status will have a high demand for LTC. Therefore, hypothesis 3 was also verified. Hoi le et al. conducted a survey of 2240 Vietnamese elderly aged 60 years or over and found out that the socioeconomic status of the elderly (such as education, household income) could affect their needs for LTC (Hoi le et al., 2011). The results of his study were similar to this study. The results of this study also explained the residents of Ontario, Canada, whose socioeconomic status were lower tended to have a higher intention of being cared and use caring services more frequently (Laporte et al., 2007). In addition, to a certain extent, the results also explicate why those from the underdeveloped areas reported more unmet care needs among the 1,251 British stroke patients (McKevitt et al., 2011).
4.4.4 Association between SES and mental health
A positive association was observed between SES and mental health in this study.
The finding is consistent with previous studies: 1) people with lower socioeconomic status tend to experience more mental health problems (Gong et al., 2012); 2) individuals from the higher socioeconomic status groups are more likely to have a better mental health (Mavrinac et al., 2009; Sani et al., 2010).
4.4.5 Discussion on the association between SES, mental health and NLTC Using the structural equation modeling, this study explored the relationship among socioeconomic status, mental health and the need for LTC. As shown in the results, socioeconomic status could not only have a direct impact on the need for LTC, but also could indirectly influence the need for LTC via mental health. It also, to some degree, shows the important role of socioeconomic status in determining the need for LTC.
Existing studies consistently showed that there is a positive correlation between the individual’s socioeconomic status and their health (including mental health), which means that a higher socioeconomic status of the individual, his health will be relatively better (Brennan & Singh, 2012; Fiorillo & Sabatini, 2011; Hwang et al., 2010; Wengler, 2011). Specifically, the difference on socioeconomic status usually means the differences of the ability on obtaining the medical services, occupational risks, social support, and life stress. These differences ultimately led to the different level of individual’s health (Kagamimori et al., 2009). Some researchers also give a similar explanation as below: those who have higher socioeconomic status are more likely to obtain medical services, more easily to build and maintain their social relationships and more likely to live in good communities, such group will eventually
123
have a higher level of health (Mirowsky et al., 2000; Robert & House, 2000). It should be emphasized that, from a macro perspective, the impact of individual’s socioeconomic status on health is not simply caused by socioeconomic status only, but caused by various factors combined together, including the individual’s lifestyle, environmental factors and the social factors and so on (Berkman & Kawachi, 2000;
Marmot & Wilkinson, 1999). Of course, there are also researchers believe that ecological factors also play an important role in the process (Kawachi & Kennedy, 1997).
The influence of socioeconomic status on the health narrated above, combining with the structural analysis of the association between socioeconomic status, mental health and the need for LTC among the older persons in this study, the process of the socioeconomic status exerting its effect on the need for LTC can be further specifically summarized as follows: those elderly who are on different socioeconomic status live in different communities, they have different lifestyle, owning different ability to obtain medical services and various degrees of social support. All of these lead to their different mental health conditions, and eventually, they have different levels of the need for LTC.