Biases in cost measurement for economic evaluation studies in health care

Abstract

This paper addresses the issue of biases in cost measures which used in economic evaluation studies. The basic measure of hospital costs which is used by most investigators is unit cost. Focusing on this measure, a set of criteria which the basic measures must fulfil in order to approximate the marginal cost (MC) of a service for the relevant product, in the representative site, was identified. Then four distinct biases--a scale bias, a case mix bias, a methods bias and a site selection bias--each of which reflects the divergence of the unit cost measure from the desired MC measure, were identified. Measures are proposed for several of these biases and it is suggested how they can be corrected.

Are Vaccination Sites in Bangladesh Scale Efficient?

“…Our evidence provides empirical support to Jacobs and Baladi's (9) contention that assuming constant returns to scale might not be realistic. The presence of increasing returns has two particular implications.…”

Assessing the costs of healthcare technologies in clinical trials.

“…Most of the methods used to generalise results are based on sensitivity analysis (Baltussen et al, 1996;Jacobs and Baladi, 1996;O'Brien, 1996). These analyses incorporate data from a variety of sources.…”

Cost-effectiveness of scaling up mass drug administration for the control of soil-transmitted helminths: a comparison of cost function and constant costs analyses

“…This analysis shows the fundamental importance taking into account economies of scale can have in economic assessments, and how ignoring this aspect, which is done in most economic assessments, can have a detrimental eff ect on con clusions, potentially incorrectly aff ecting policy. 28 This fi nding is particularly relevant to NTDs because of the nature of the costs of MDA-many of which are fi xed, because the drugs themselves are often donated or inexpensive-but is also relevant to other large-scale control programmes. 13 This research generated two other important conclusions for resource allocation and programme planning.…”