Examining variations in hospital productivity in the English NHS

Adriana Castelli, Andrew David Street, Rossella Verzulli, Padraic Ward

Research output: Contribution to journalArticlepeer-review

Abstract

Objectives
Numerous papers have measured hospital
efficiency, mainly using a technique known as data
envelopment analysis (DEA). A shortcoming of this technique
is that the number of outputs for each hospital generally
outstrips the number of hospitals. In this paper, we
propose an alternative approach, involving the use of
explicit weights to combine diverse outputs into a single
index, thereby avoiding the need for DEA.


Methods
Hospital productivity is measured as the ratio of
outputs to inputs. Outputs capture quantity and quality of
care for hospital patients; inputs include staff, equipment,
and capital resources applied to patient care. Ordinary least
squares regression is used to analyse why output and productivity
varies between hospitals. We assess whether
results are sensitive to consideration of quality.

Results
Hospital productivity varies substantially across
hospitals but is highly correlated year on year. Allowing for
quality has little impact on relative productivity. We find
that productivity is lower in hospitals with greater financial
autonomy, and where a large proportion of income derives
from education, research and development, and training
activities. Hospitals treating greater proportions of children
or elderly patients also tend to be less productive.

Conclusions
We have set out a means of assessing hospital
productivity that captures their multiple outputs and
inputs. We find substantial variation in productivity among
English hospitals, suggesting scope for productivity
improvement.

Original languageEnglish
Pages (from-to)243-254
Number of pages12
JournalThe European Journal of Health Economics
Volume16
Issue number3
Early online date24 Feb 2014
DOIs
Publication statusPublished - Apr 2015

Keywords

  • Hospital Sector
  • Productivity
  • Production functions
  • DEA

Cite this