The evaluation of health care

This paper is about the practical problems of evaluating health care. It argues that too little attention has been paid to the how and why, the what and the wherefore of evaluation. First, the paper reviews the well known problems of the classic evaluation model: in particular the difficulty of generalising the findings of a randomised control trial (RCT) to real clinical situations and the concentration of cost-effectiveness analyses (CEA) on the margins of the health sector where criteria and objectives can be debated or where there is a question of relative exposure and power. The second section examines the problem of measuring output in three typical situations: the comparison of different techniques for screening diagnosis, maintenance or therapy; the measurement of the cost-effectiveness of instructions in providing health care; and the operative evaluation of the health care system as a whole. The third section examines the sociological approach to evaluation in the same three typical situations: a comparison of different treatments; the effectiveness of antenatal care; and the implementation of health policy. Whilst these evaluations are considerably more sensitive, they still ignore some of the crucial questions. The final section considers the fundamental problem of participation and power. It argues that all evaluations are ideologically based even a thorough going anthropological approach being open to bias and distortion. It concludes in favour of the alternative of popular participation and control whilst recognising the considerable organisational and political problems involved.