Informal Care and Medical Care Utilization in Europe and the United States Alberto Holly 1, Thomas M. Lufkin 1, Edward C. Norton 2, Courtney Harold Van Houtven 3 Prepared for the Workshop on Comparative International Research Based on HRS, ELSA and SHARE RAND, Santa Monica, CA July, 10-11, 2006 1 Institute of Health Economics and Management, University of Lausanne 2 Department of Health Policy and Administration, University of North Carolina at Chapel Hill 3 Center for Health Services Research in Primary Care, Durham Veterans Affairs Medical Center and Department of Medicine, Division of General Internal Medicine, Duke University Medical Center
Outline Introduction Institutional settings Results for the United States Results for Europe Discussion Informal Care and Medical Care Utilization in Europe and the United States 2
Introduction Definitions Research questions Usual discussion: Separation medical and long term care Reduce the importance of institutional care with respect to home care for two main reasons Preferred mode Less costly (also in relation to the forecast in demographical changes. Informal Care and Medical Care Utilization in Europe and the United States 3
Introduction Our approach is systemic: A systems view of the health sector is useful in trying to understand health care and LTC finance Emphasis on substitution effect and complement effect Informal care by adult children is a common form of long-term care for older adults and can reduce medical expenditures if it substitutes for formal care Informal Care and Medical Care Utilization in Europe and the United States 4
Introduction We build on work by Van Houtven and Norton (2004) Compare results from Europe with the SHARE data to results from the United States with the HRS/AHEAD data. Highlight the main institutional differences between Europe and the United States as a motivation for our conceptual framework Importance of endogeneity issue Informal Care and Medical Care Utilization in Europe and the United States 5
Institutional settings A detailed description is given in OECD (2005) European countries offer some public provision of health care, either through a national health service, a social insurance scheme or a compulsory private insurance system. All of these systems include some long-term care as part of the basic coverage. In general, provision of care is only dependent on need and not on income or wealth. Supplementary or complementary insurance schemes provide additional coverage and allow for more use of services. Informal Care and Medical Care Utilization in Europe and the United States 6
Institutional settings In the United States the Medicare but offers limited home health and nursing home care. There is no universal right to public services in the United States, The share of individuals with private longterm care insurance remains small. Informal Care and Medical Care Utilization in Europe and the United States 7
Institutional settings Long term care + Medical care Home care Formal care + Informal care Institutional care Informal Care and Medical Care Utilization in Europe and the United States 8
Results for the United States Prior studies of the relationship between informal and formal care in the United States find mixed results. They did not seem to definitively establish whether the two types of care were substitutes or complements. The few studies that control for endogeneity (Lo Sasso and Johnson, 2002, Greene, 1983) suggest that informal care and formal care are substitutes. Van Houtven and Norton (2004) study the impact of informal care by children on formal care utilization: Informal Care and Medical Care Utilization in Europe and the United States 9
Results for the United States Their sequential decision model: The children first determine their optimal level of informal care provision, given their budget and time constraints and their parent's health, Then parents choose optimally their formal care consumption, given their health level and the available supply of informal care and subject to their budget constraint. Informal Care and Medical Care Utilization in Europe and the United States 10
Results for the United States The formal model is then: Mijt = git (Ait, Hit, Xit, Cit, εijt) Mijt, the parent's utilization of formal care, is a function of informal care Ait, health Hit, demographic and income characteristics Xit, and other consumption goods Cit such as insurance. Informal Care and Medical Care Utilization in Europe and the United States 11
Results for the United States The function g is either the probit function or is linear for the second part of the two-part model, depending on the availability of data for each type of formal care. Hypothesis: the variable of interest, hours of informal care provided by the children, Ait, is endogenous They instrument it with children characteristics, such as the number of children, especially of daughters. Informal Care and Medical Care Utilization in Europe and the United States 12
Results for the United States Using data from the (HRS) and the Asset and Health Dynamics Among the Oldest-Old Panel Survey (AHEAD), they find that: Informal care Reduces home health care use Delays nursing home entry, They detect endogeneity for all utilization types, but outpatient surgery. Informal Care and Medical Care Utilization in Europe and the United States 13
Results for Europe In our study, given the structure of the SHARE questionnaire, the analysis is restricted to individuals who are aged above 50, live alone (as singles, divorcees, widows) are not institutionalized and have at least one child. Informal Care and Medical Care Utilization in Europe and the United States 14
Results for Europe Bolin, Lindgren and Lundborg (2006) applying the same model as Van Houtven and Norton s (2004), and using the SHARE dataset, showed that Informal and formal home care are substitutes, But that informal care appears to be a complement to other types of formal care. They also detected endogeneity in the case of home care, while this was not the case for the other types of formal care. Informal Care and Medical Care Utilization in Europe and the United States 15
Results for Europe Our model: The modeling of formal and informal care should take into account the institutional differences between Europe and the United States We suppose that parents first make their decision to use formal care and then the quantity is decided according to their medical need. Informal Care and Medical Care Utilization in Europe and the United States 16
Results for Europe Children then decide whether they will provide informal care to their parents or not, depending on the health state of the parent and the amount of formal care received, subject to their time and budget constraints. We thus model the parent s utilization of informal care as: I i =f(qf i, H i, S i, C i, ε i ) Informal Care and Medical Care Utilization in Europe and the United States 17
Results for Europe Where: QF i is the quantity of formal home care. This variable is assumed to be endogenous H i is a measure of health S i socio-economic and demographic characteristics C i children characteristics The function f is a probit function because we use a binary measure of informal care I i Informal Care and Medical Care Utilization in Europe and the United States 18
Results for Europe Instruments: Objective measures of health and behavior as instruments, such as the number of severe conditions (e.g. cancer, heart problems, diabetes) a patient has, Some indicators of his or her health behavior (whether he or she is a former smoker and the level of physical activity) The number of years of education of the parent. Informal Care and Medical Care Utilization in Europe and the United States 19
Results for Europe We also use a simultaneous equations approach (bivariate probit) to model the relationship between formal and informal care allowing for the endogeneity of formal home care: Ii = f( Fi, Hi, Si, Ci, εi) Fi = g( Hi, Si, Ci, ηi) Where F i is a binary measure of formal care utilization Informal Care and Medical Care Utilization in Europe and the United States 20
Results for Europe Variables Informal care: dummy variable taking the value 1 if the respondent had received any type of care and zero otherwise (No distinction in the type of informal care received). Formal care: two variables Binary variable Number of hours that the respondent has received either professional nursing care or home help Informal Care and Medical Care Utilization in Europe and the United States 21
Results for Europe Other explanatory variables divided into three categories. Socio-economic characteristics of the respondent, such as age, education, wealth and income and insurance status. Health variables such as the number of ADL and IADL or self perceived health. Variables concerning the children as informal caregivers (e.g., number of children and the number of children living more than 100 Kilometers away from the respondent). Informal Care and Medical Care Utilization in Europe and the United States 22
Variable name Description informal dummy=1 if individual receives informal care by his/her children or grandchildren loghomecare ln(total number of hours of formal care+1) age age of individual male dummy=1 if individual is a male yearsedu number of years of education incomecat income category (min=1, max=6) wealthcat wealth category (min=1, max=6) insurance dummy=1 if individual has complementary and/or supplementary insurance adl number of ADL limitations iadl number of IADL limitations sphus self perceived health, US version proxy dummy for proxy respondent currentsmoker dummy=1 if respondent is currently a smoker formersmoker dummy=1 if individual is a former smoker Informal Care and Medical Care Utilization in Europe and the United States 23
Variable name nconditions drinkin2 activities nbchildren childagemax chinherit cho100km Description number of conditions (heart, stroke, hip fracture, falls, diabetes, cancer, lung, arthritis, incontinence, eyesight) dummy=1 if drinking > 2 glasses of alcohol almost every or 5/6 days a week frequency of engagement in activities that require low or moderate level of energy (gardening, cleaning the car, doing a walk...) (min=1, max=4.:1=more than once a week, 2=once a week, 3=one to three times a month, 4=hardly ever, or never) number of children (min=1, max=17) age of eldest child percent chance of leaving an inheritance of more than 50,000 Euro number of children leaving over 100 Km away (min=0, max=4) Informal Care and Medical Care Utilization in Europe and the United States 24
Table 2: Summary statistics of the variables Variable Obs Mean Std. Dev Min Max informal 3892 0.2207091 0.414778 0 1 loghomecare 3892 0.5054185 1.604833 0 16.117 age 3892 69.61074 11.24059 50 102 male 3892 0.2317575 0.4220091 0 1 yearsedu 3892 9.220324 4.559293 0 22 incomecat 3892 2.625899 1.516406 1 6 wealthcat 3892 2.681912 1.633324 1 6 insurance 3892 0.2502569 0.4332166 0 1 adl 3892 0.3044707 0.9145167 0 6 iadl 3892 0.5704008 1.265125 0 7 sphus 3892 3.125128 1.076377 1 5 proxy 3892 0.0629496 0.2429036 0 1 nconditions 3892 3.640802 1.63361 1 11 drinkin2 3892 0.0760534 0.2651177 0 1 currentsmoker 3892 0.1929599 0.3946725 0 1 formersmoker 3892 0.204779 0.4035919 0 1 activities 3892 1.72148 1.134049 1 4 nbchildren 3892 2.397739 1.406222 1 13 childagemax 3892 38.66675 11.77986 5 76 Informal Care and Medical Care Utilization in Europe and the United States 25
Table 3: Estimation results: biprobit (1) informal homehctot sphus proxy drinkin2 currentsmoker formersmoker nbchildren childagemax chinherit cho100km male adl iadl nchronic incomecat wealthcat insurance age yearsedu rural Constant (2) homehctot -0.946 (2.63)** 0.143 (5.02)** 0.179 (4.98)** 0.17 (1.32) -0.372 (2.86)** -0.462 (3.86)** -0.057 (0.45) 0.175 (2.36)* 0.347 (3.90)** 0.055 (0.82) 0.078 (0.91) 0.043 (2.35)* 0.002 (0.07) 0.017 (3.77)** 0.008 (1.33) 0.002 (3.37)** 0 (0.13) -0.169 (4.15)** 0.046 (1.15) -0.194 (2.79)** 0.027 (0.32) 0.002 (0.05) 0.002 (0.06) 0.193 (5.90)** 0.229 (7.39)** 0.024 (1.17) 0.085 (4.01)** 0.007 (0.36) 0.027 (1.11) -0.063 (3.10)** -0.078 (2.97)** 0.188 (2.37)* 0.544 (7.79)** 0.016 (2.86)** 0.035 (5.62)** 0.012 (1.95) 0.014 (1.77) -0.026 (0.39) -0.235 (2.74)** -3.313 (9.88)** -5.425 (15.76)** (3) Intercept (Rho) 0.779 (0.380)* Observations 3595 3595 Absolute value of z statistics in parentheses * significant at 5%; ** significant at 1% Informal Care and Medical Care Utilization in Europe and the United States 26
Table 4: Estimation results: IVprobit informal. loghomecare 0.193 (2.05)* age 0.005 (0.86) male -0.214 (3.31)** incomecat -0.022-1.13 wealthcat -0.056 (2.80)** adl -0.033 (0.96) iadl 0.033 (0.80) sphus 0.102 (3.96)** proxy 0.179 (1.70) drinkin2-0.477 (4.10)** currentsmoker 0.084 (1.23) nbchildren 0.049 (2.76)** childagemax 0.015 (3.24)** chinherit 0.003 (4.83)** cho100km -0.18 (5.11)** Constant -2.145 (7.73)** Observations 3884 Absolute value of z statistics in parentheses * significant at 5%; ** significant at 1% Informal Care and Medical Care Utilization in Europe and the United States 27
Discussion For the European countries: apparent contradictory results regarding formal and informal care: Bivariate probit model: could be interpreted as a substitution effect; IVprobit model: the provisions of informal and formal care are complements Possible explanation? Informal Care and Medical Care Utilization in Europe and the United States 28
ρ Discussion Hypothesis to be tested in a future version: Informal care is a substitute to formal care for low users or individuals with low needs, for parents with very high needs, then informal care comes on top of formal home care, as reflected by the positive sign of ρ This is similar to the findings of Van Houtven and Norton (2006) that informal care reduces the probability of using formal care, but does not reduce the level of expenditure on home care, once long-term care is accessed. Informal Care and Medical Care Utilization in Europe and the United States 29
ρ Discussion Taking the institutional setting into account makes it difficult to compare directly the relationship between informal care and formal care in Europe and in the United States. A common feature: endogeneity of informal care. Informal Care and Medical Care Utilization in Europe and the United States 30
Discussion For Europe, Utilization of home health care leads to a decrease in the probability of receiving informal care provided by children, But an increase in the number of hours of home care received tends to increase the probability of receiving informal care, While in the United States, informal care is found to be a substitute to home health care. Informal Care and Medical Care Utilization in Europe and the United States 31