Private Long-Term Care Insurance (LTCI) and Patterns of Care Use among Older Adults

Private Long-Term Care Insurance (LTCI) and Patterns of Care Use

among Older Adults

Yong [email protected]

AcademyHealthJune 10, 2008

Authors

Yong Li, Ph.D, School of Public & Environmental Affairs, Indiana University Purdue University – Indianapolis (IUPUI)

Gail A. Jensen, Ph.D, Department of Economics and Institute of Gerontology, Wayne State University

Background

Long-term care financing

Source: Congressional Budget Office, 2004

Shares of spending on long-term care for the elderly, 2004

Growing interest in LTCI

Long-term care is not adequately insured• High cost, misperception, Medicaid…

Growing interest among researchers and policy makers.• Welfare model insurance model

Little is known about how LTCI may change the pattern of care use.

Research objectives

We examine the effects of LTCI on patterns of care use among disabled older adults ages 65 and over.

We extend earlier work in three ways.

• Look at three major types of long-term care

• Address the issue of endogeneity

• Use most recent and nationally representative data from HRS

Design and Methods

Data

Pooled HRS data wave4 – wave7 Study sample:

• Older adults who were 65 years old or older• Reported limitations in at least two activities

of daily living (ADLs).• Sample size = 4,319 (2,711 unique

individuals) Robust standard errors are reported in all

regression analyses.

Empirical specification

LTC: use of nursing home care, formal home care, or informal care

INS: whether an individual has LTCI• INS is presumed to be correlated with ε

X: a vector of exogenous covariates g(): probit or linear

Two-part models of utilization

Skewed distribution in the dependent variables• Significant occurrence of the value 0• OLS inappropriate

A series of two-part models are implemented

Part I: probit model of any positive use. Part II: linear model of the amount of care,

conditional on having any.

Endogeneity of LTCI

LTCI may be subject to the endogeneity problem.• OLS inconsistent• No causal interpretation

We formally test for endogeneity of LTCI in each model estimated.

Identifying variables:• Whether one has a life insurance policy.• Price (premium) of LTCI policy.

Results and Discussion

Main findingsTable 3: Effect of private long-term care insurance on care utilization

Model Coef. (S.E) Marginal effect

Nursing home care

Pr(any nursing home care) -0.190 (0.119)

E(log nights in nursing home | nights>0) -0.124 (0.095)

E(nights in nursing home) -28.461

Formal home care

Pr(any formal home care) 1.540** (0.649)

E(log hours of formal home care | hours>0) -1.786 (1.528)

E(hours of formal home care) 64.995

Informal care

Pr(any informal care) 0.071 (0.100)

E(log hours of informal care | hours>0) -0.009 (0.140)

E(hours of informal care) 2.745

Effect of LTCI on nursing home care utilization

Estimation sample N Part I Part II

Full sample (2-5 ADL limitations) 4,319 -0.190 (0.119) -0.124 (0.095)

4 or 5 ADL limitations 1,788 0.037 (0.157) -0.035 (0.093)

2 or 3 ADL limitations 2,531 -0.547** (0.236) -0.638** (0.262)

Note: ** Significant at the 5% level.

Robust standard errors in parenthesis

Conclusions LTCI significantly changes the pattern of care

utilization among older adults.• Avoid or at least postpone entering a nursing home.• Receiving care in their preferred setting.• No evidence of informal care cut back.

LTCI is a more efficient way to finance LTC services.• Directs resources to formal home care, without

discouraging family care-giving.• Reduces unnecessary nursing home stays