Abstract
Introduction:
Levels of management staffing in the UK National Health Service (NHS) have received considerable political and media attention in the last four years or so. Both the previous and current governments committed themselves to reducing management expenditure in the health service, and significant cuts appear to have been made in this area in the last year or so. Few systematic evaluations have been undertaken of the value of general management input, however, and policy changes appear to have been largely determined by popular opinion. This study attempts to quantify the effect of management input on hospital productivity for the first three years of the NHS internal market (1991/2–1993/4).
Methods:
An average cost function was used to model the effect of management inputs on hospital costs after adjustment for the levels of outputs produced, input costs, and internal and external exogenous constraints on hospital functioning. Two measures of management input were used: the proportion of total spending consumed by top level management, and the proportion consumed by all administrative activities. Cross-sectional and longitudinal models with contemporaneous and lagged management input effects were estimated.
Results:
Higher spending on top level management was associated with poorer productivity levels in most instances. Total administrative inputs had a weaker, but still generally negative, association with productivity.
Conclusions:
No evidence was found that increasing management inputs was associated with improved productivity. On the contrary, spending more on top level managers appeared in fact to be associated with lower productivity levels. Results would thus appear to be in line with decisions to reduce the level of expenditure on management in NHS hospitals. Quality differences could not be measured, however, and it is possible that management input is associated with quality improvements which might reduce or reverse observed productivity losses.
Get full access to this article
View all access options for this article.
