XXXIV Simposio Análisis E conómico A double sample selection model for unmet needs, formal care and informal caregiving hours of dependent people in Spain Cristina Vilaplana Prieto (UMU, FEDEA) Sergi Jiménez-Martín (UPF, FEDEA) Valencia, December 2009
29
Embed
XXXIV Simposio Análisis Económico A double sample selection model for unmet needs, formal care and informal caregiving hours of dependent people in Spain.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
XXXIV Simposio Análisis Económico
A double sample selection model for unmet needs, formal care and informal caregiving hours of
dependent people in Spain
Cristina Vilaplana Prieto (UMU, FEDEA)
Sergi Jiménez-Martín (UPF, FEDEA)
Valencia, December 2009
XXXIV Simposio Análisis Económico
Motivation
Factors affecting the supply of informal caregiversAverage household size:
Percentage of mentally ill people among those receiving informal care
1994 19.2%2004 39.6%(Informal Support Survey, 1994 y 2004)
Population projections
XXXIV Simposio Análisis Económico
Motivation
Preferences of elder people and informal caregiversOpinion about the participation of Public Sector in long-term care
Preference for the place for receiving care in case of need
XXXIV Simposio Análisis Económico
Motivation Previous research has concentrated on: causes and
consequences of unmet needs (UNs) (Allen, 1994; Tennstedt et al.,
1994; Desai et al., 2001; Sands et al., 2006 ). Consequences of UNs: Always negative
Morbility rate Degree of severity of disabilitiesDownfalls and accidents Use of emergendy departments Physician visits Hospital stays Institutionalization rate
Health expenditure(Allen and Moor, 1997)(Komisar et al., 2005; )(Long et al., 2005)
Limiations of previous studies:Do not take into accout the characteristics of informal caregivers network Omission of principal caregiver opinion
XXXIV Simposio Análisis Económico
Motivation Reliability of informal caregivers’ reports as a valid
instrument for measuring the number of informal hours and the type of services received (Williams et al., 1997 ; Albert et al., 2004 )
Spanish literature about UNs: scarce
XXXIV Simposio Análisis Económico
Questions
To what extent personal social services for dependent people (Home Care and Day Centres) are able to satisfy dependent’s needs
What is the effect of formal unmet needs over informal caregiving hours?
Which factors are associated to the emergence of personal unmet needs?
XXXIV Simposio Análisis Económico
Data Informal Support Survey (IMSERSO, 1994 & 2004) Method: household interview Who answers: informal caregivers (relatives, friends,
neighbours) of older people (60 years and older), who may be co-resident or not
Sample: N=1.665 in 1994, N=1.504 in 2004 Ceuta, Melilla and La Rioja excluded
Conclusions can not be applied to: Institutionalized dependent people Individuals who only receive formal care
Innovative approach for determining the dependency degree Application of the Ranking Scale of Dependency (R. D. 504/2007, de 20 de abril)
XXXIV Simposio Análisis Económico
Descriptive statistics
Classification in three dependency degrees:Moderate: needs help for DLA one a daySevere: needs help two or three times per dayGreat: needs help for DLA several times per day
Two levels of supervision inside each degree:Level 1: can perform the activity by himselfLevel 2: requires some degree of support
Compare the survey questionnaire with the official one47 tasks grouped in 10 activities (+ an additional activity for mentally ill/cognitive disabled)
25-49: Moderate; 50-74: Severe; 75-100: Great
XXXIV Simposio Análisis Económico
Descriptive statistics
Slight increase in the percentaje of no-dependent people as a consequence of healthy life expectancy High degree of consistency between the classification obtained form Informal Support Survey (1994, 2004) and the Disabilities, Deficiencies and Health Status Survey (1999)
XXXIV Simposio Análisis Económico
Definition of Unmet Needs (UN)
Not satisfied (Quantity, quality)
Unmet needs Problem
Rejected Accepted
Dependent(or informal caregiver)
applies for public formal care
Satisfied
Home CareDay Centre
XXXIV Simposio Análisis Económico
Descriptive statistics
Relation betwen FC and UN according to dependency degreeComparison 1994/2004
Decrease in 2004Higher for great dep.
Slight increase in 2004
XXXIV Simposio Análisis Económico
Double sample selection model
Relation between FC and UNs:
difference between the amount of services offered and the conditions required to be awarded.
difference between the expected benefit from formal care and the current provision of services.
are unobsevable
XXXIV Simposio Análisis Económico
Double sample selection model
In the hours equation we must take into account
Define an equation for each subsample where we include the selectivity terms to correct the double selection problem (Tunali, 1986):
’s: Selectivity terms
XXXIV Simposio Análisis Económico
Explanatory variables
Carerecipient’s characteristicsGender, sex, marital status, level of education, dwelling arrangements, degree of dependency, pathologies, type of benefit received, monthly household income
Caregiver’s characteristicsGender, sex, level of education, marital status, number of caregiving years, permanent caregiver, wilful caregiver, kinship dependent-caregiver, good relation with dependent before the onset of caregiving tasks
Informal network:Secondary caregivers, kinship primary-secondary caregiver, children living at household
Formal carePrivate formal careSocial services: Home Care (coverage index, time devoted to personal care, hours/week, cost/hour, copayment), Day Centre (coverage index, % psycogeriatric places, copayment)
Identification of equationsNot included in FC equationNot included in UN equationUnderlined not included in Hours equation
Size of muncipality
XXXIV Simposio Análisis Económico
Bivariate probit model: Results
Correlation coefficient NegativeDependent people who receive FC are less prone to suffer UNBigger effect in 2004
Marginal effects for binary variables using the sample mean and the sample median
Pathologies
Almost constant for dementia osteoarticular problems
XXXIV Simposio Análisis Económico
Bivariate probit model: Results
Great dependent: FC=0&UN=1 by 47.01% in 1994 and 22.10% in 2004. Moderate dependent:
Lower decrease in the FC=0&UN=1 than GDDifferences between dependency degrees have shortened.
Severe or Great dependent: FC=1&UN=1 by 12.54% and 25.18% in 1994 and has
raised to 22.33% and 35.44, respectively, in 2004 higher probability of both receiving FC and considering it is unsatisfactory. Moderate dependency:
If they receive FC, lower prob. of not being satisfied
XXXIV Simposio Análisis Económico
Bivariate probit model: Results
Higher coverage index of Home Care/Day Centre FC=0&UN=1 & FC=1&UN=1In 2004, the negative effect over: FC=0&UN=1 is higher for Day Centre
FC=1&UN=1 is higher for Home Care
XXXIV Simposio Análisis Económico
Bivariate probit model: Results
Home Care includes personal care and houseworkingMore time devoted to personal care: FC=1&UN=1
High variability across regions: Maximum: Navarra 80% Minimum: Extremadura 20%
More hours/month FC=0&UN=1 FC=1&UN=1 High variability across regions:
Maximum: Galicia 25.14 hours/month Minimum: Andalucía 8 hours/month
XXXIV Simposio Análisis Económico
Bivariate probit model: Results
Cost/hour & copayment: FC=1&UN=1 and FC=0&UN=1 Cost/hour:
1994:22>0, 42<0 FC=1&UN=1more IH than FC=0&UN=1 2004:
21<0: FC=0&UN=1 more IH than FC=0&UN=0 31>0: FC=1&UN=0 less IH than FC=1&UN=1 42<0: FC=1&UN=1more IH than FC=0&UN=1
FC=0UN=0
FC=0UN=1
< FC=1UN=1
< > FC=1UN=0
XXXIV Simposio Análisis Económico
Informal Caregiving Hours: Results
Higher coefficients for FC=1&UN=1 as compared to FC=0&UN=1 .
XXXIV Simposio Análisis Económico
Informal Caregiving Hours: Results
For the same dependency degree, informal caregivers devote more hours in 2004 than in 1994. For example, a level 2 great dependent with FC and UN implied an increase of 4.65 hours/week in 1994 and 5.64 hours/week in 2004. .
XXXIV Simposio Análisis Económico
Informal Caregiving Hours: Results
The distance in caregiving hours between moderate (level 2) and great dependence (level 2) has increased between both waves:
2.76 hours/week in 1994 3.62 hours/week in 2004.
XXXIV Simposio Análisis Económico
Informal Caregiving Hours: Results
Difference between with/without UNFC=0&UN=0 vs. FC=0&UN=1
1994 1.09 more hour/week2004 1.68 more hour/week
FC=1&UN=1 vs. FC=1&UN=01994 1.70 more hour/week2004 2.01 more hour/week
XXXIV Simposio Análisis Económico
Predicted number of IC hours
Wives provide more care than daughters for the national average Significant differences by Autonomous Communities related to the
provision of social services
XXXIV Simposio Análisis Económico
Conclusions
Negative relationship between FC & UN 1994: self-selection in UNs equation 2004:
Significant self-selection in both equations Selection in UN eq higher than in FC eq Evidence against Substitution Model (FC IC) Dependent’s variables not considered in conditions required
for receiving FC in 2004 Significant effect over Informal Hours
Consequences of UNs extended to informal caregiver’s life
Home Care: formal caregivers should... Assume that althoug IC is available, dependent people do not want
to overburden family members Homogenisation of the % of time devoted to houseworking and
personal care across Autonomous Communities
XXXIV Simposio Análisis Económico
Future research
New Law of Dependency Individual Care Program (art. 29): “the modes of
intervention that are most suitable to their needs from among the services and financial benefits foreseen”
Consideration of insufficient coverage: “The services in the Catalogue shall be of a priority nature and shall be provided through the Social Services Network by the respective Autonomous Communities or subsidised private centres or services” (art. 14.2)
Is there a significant reduction in the percentaje of UNs?
Are some UNs more important than others? What is more important in terms of dependent’s quality of life: the
elimination of a moderate UN or the reduction of a severe one?