Applying a conventional Data Envelopment
Analysis (DEA) and the homogeneous bootstrap procedure, we found that the
efficiency levels for both public and private nursing homes in Ireland do not
exceed 50 per cent on average – a finding which corroborates results obtained
for the nursing home sector in other countries. Furthermore, given the observed very low
efficiency scores, this paper aims to estimate the relationship between
efficiency and its possible determinants.
We use a comprehensive set of environmental variables in 59 public and
93 private Irish nursing homes for the period 2008-2009. As well as ownership and other characteristics
such as size, age, regulatory constraints and business
environment, we examine how the quality of care and the case mix
affect the efficiency scores of the nursing homes in Ireland. Many previous studies have employed a
two-stage approach wherein nonparametric DEA efficiency estimates
from the first stage are regressed on a vector of some environmental variables
in a
parametric analysis which most often specified a censored (Tobit)
model for the second stage. The
typical two-stage studies do not provide a coherent description of the underlying
data-generating process (DGP), and the method of inference is flawed since the
DEA efficiency estimates are biased and are serially correlated. Whatever the second-stage regression
specification employed, conventional inference methods fail to give valid
inference due to the fact that in the second-stage, true efficiency remains
unobserved and must be replaced with DEA estimates of efficiency, and these are
correlated by construction. The
efficiency score is a point estimate without a probability distribution around
it as required by the Tobit methodology or any other parametric regression technique. Using the DEA point estimates in a second
stage analysis may cause biased and inconsistent estimates of the parameters of
the environmental variables. We apply a double bootstrap DEA
approach in order to obtain unbiased and consistent results. We show that ownership does not have a
significant effect on technical efficiency, but the size of the nursing home
has a positive influence. More
importantly, the quality of care and the case-mix are important factors
determining efficiency scores in Irish nursing homes. In terms of policy
implications, our results suggest that whereas quality, as indicated by the
proportion of single beds, decreases technical efficiency, appropriate training
for medical staff such as a diploma in gerontology should be incentivised.