Reducing Nursing Home Use Through Community Long-Term Care: An Optimization Analysis Using Data from the National Channeling Demonstration. Scope of the Study


This study explores the capability of community long-term care (CLTC) services to reduce nursing home use when services are allocated strategically for this purpose. We accomplish this by first considering an actual CLTC clientele--the persons screened into the National Long-Term Care Channeling Demonstration (NLTCCD). We take note of these persons' existing use of CLTC services and their costs, as well as the nursing home use experience of the population. We then through simulation reallocate the existing budget for these services over the population so that the service packages received by each individual have the property that projected total nursing home use by the population is minimized.

In this way, we begin to answer some key questions regarding CLTC services that have not yet been systematically addressed in the literature. First, by comparing nursing home use under the simulated optimum service allocation with that actually observed, we assess the technical capacity of CLTC services to reduce nursing home use when optimally committed to this goal. Second, by comparing the existing allocation of budget and services in the NLTCCD population with that which minimizes nursing home use, we gain insight into the nature of existing inefficiencies and possible changes in aggregate service mixes and individual use criteria that may be broadly useful in policy analysis and service planning.

Technically, the problem solved here is a large-scale nonlinear programming (NLP) problem. We use a logistic transition-probability model (TPM) to empirically estimate the relationship between use of community services and nursing home use. From these results, we parameterize an objective function for the optimization problem, which we then solve.

The plan of the study is this. We begin by briefly reviewing the relevant literature on community services and nursing home use. We then define a transition probability model for nursing home use and detail the variables in terms of which it is estimated. Next, we define our optimization model, and link it to the TPM. Finally, we describe our sample, present the empirical results from the estimation of the TPM and the consequent NLP solution, and discuss their policy implications.

We should note that we are not concerned here specifically with the effects of the Channeling demonstration. Rather, we are interested in the broader question of the influence of formal CLTC services as such, regardless of their source and method of financing. Thus we treat the additional services provided by the demonstration just as those from any other source, public or private.

We emphasize also that the optimization that we conduct assumes that all existing expenditures in the study sample for community services are under its control, whether services are financed by public or private sources. Implied is a rather authoritarian "social planning" model in which public resources can be reallocated at will, and private resources can in effect be taxed away when they can be more effectively used elsewhere in reducing nursing home use in the community as a whole. We do not endorse this model as desirable or, as a practical matter, feasible. The issue we seek to isolate and address clearly is the theoretical potential of a fixed budget for community services to reduce nursing home use when services are unequivocally committed to this as a goal. Once this baseline potential has been established, it will then be possible to better understand the consequences of the constraints, inefficiencies, and multiple purposes found in the real world. On the other hand, if the potential to reduce nursing home use is found to be negligible even under conditions which theoretically maximize this effect, then this outcome can safely be abandoned as a policy goal.

View full report


"rednh.pdf" (pdf, 1.92Mb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®