Developing new approaches to measuring NHS outputs and productivity

The Centre for Health Economics and National Institute of Economic and Social Research have recently completed a project funded by the Department of Health to improve measurement of the productivity of the NHS. The researchers have suggested better ways of measuring both outputs and inputs to improve estimates of productivity growth. Past estimates of NHS output growth have not taken account of changes in quality. The CHE/NIESR team conclude that the routine collection of health outcome data on patients is vital to measure NHS quality. They also propose making better use of existing data to quality adjust output indices to capture improvements in hospital survival rates and reductions in waiting times. With these limited adjustments the team estimate that annual NHS output growth averaged 3.79% between 1998/99 and 2003/04.The research team has also developed improved ways of measuring NHS inputs, particularly by drawing on better information about how many people are employed in the NHS and by recognising that staff are becoming increasingly better qualified. There have been substantial increases in staffing levels, pharmaceutical use and investment in equipment and buildings since 1998/99. The net effect of this growth in both outputs and inputs is that, according to the research team’s estimates, NHS productivity declined by about 1.59% a year since 1998/99. This is not out of line with estimates of growth rates in other UK and US service sectors, including insurance and business services. Nor is it surprising that recent years have seen negative growth in the NHS. There are at least two reasons. First, there has been an unprecedented increase in NHS expenditure. The NHS has had to employ more staff to meet the requirements of the European Working Time Directive and hospital consultants and general practitioners, in particular, have benefited from new pay awards.Second, the NHS collects very little information about what actually happens to patients as a result of their contact with the health service. Until there is routine collection of health outcomes data, measurement of the quality of NHS output will remain partial and productivity growth is likely to be underestimated.