With the current demographic shift being experienced by populations globally, almost linear increases in life expectancy have been seen and can be expected. However, increases in healthy life expectancy may not keep pace. Among older populations the proportion of time spent in less than full health tends to increase. As a result, the accurate valuation of life spent in states less than full health will become increasingly important. Different techniques and approaches have been used to measure health in populations. The use of summary measures of population health such as DALYs (Disability Adjusted Life Years) has become common, and is widely used to compare health between populations and to evaluate the potential impact of interventions in economic analyses. Most of the commonly used summary measures of health express some measure of life lived in full health and life lived with disability or in a state of sub-optimal health. Critical to the construction of summary health measures are values assigned to health states. Current tools used in determining these values include the standard gamble, time trade off, person trade off, and the visual analogue scale. However, these techniques all have the disadvantage of incorporating individual biases (derived from particular characteristics specific to individuals or populations) into the process through which health state valuations are derived. As a consequence health states are often not directly comparable between populations, since characteristics such as nationality and ethnicity can influence how health states are valued. Furthermore, health can be judged differently by those of different ages, with the young often assigning a lower value to life lived at less than full health compared to older people. The challenge of obtaining opinions which are not influenced by an individual's own circumstances is not new. This issue was encountered and described by the American philosopher John Rawls in 'A Theory of Justice' (1971), in which he employed a thought experiment called 'the veil of ignorance' as a means of overcoming this problem. In this thought experiment an individual is asked to make decisions about distributive justice by imagining they are behind a 'veil of ignorance', whereby they are unaware of their own position in society. Here we discuss how current methods for deriving health state values may incorporate a veil of ignorance approach, and how this may benefit the comparability of the health state valuations produced. We also propose how such methods may be operationalized. Considering these issues, we propose that a new society with new needs and a progressively growing interest in maintaining adequate health requires appropriate measures of health. These measures should facilitate derivation of objective measures of health that are comparable to those acquired in other populations, irrespective of age, gender, disease status, ethnicity and geographical location. Promoting and improving health demands adequate measures of health and the application of the Rawlsian veil of ignorance approach could be an effective alternative.