The Patient-generated Index (PGI) is a health-related quality of life (HRQoL) measure which asks respondents to nominate the areas of their lives which are most affected by their health condition, so that they can then rate the severity of the effects and weight their relative importance. It is unusual amongst such measures in that it is designed for postal administration. This study assessed the ability of the revised PGI to measure change in HRQoL in a population of 161 people who had previously been identified as having limiting long-term illness. A questionnaire, including a revised version of the PGI and the developmental version of the SF-36, was mailed at two time points (T1 and T2), 4.5 months apart. The PGI was subsequently assessed in terms of practicality, validity, reliability and responsiveness. At T1, 62% of those who felt that they still had a health problem affecting their life completed the PGI correctly. These people were significantly younger and had spent longer in education than the remaining 38%. Only 19 respondents completed the PGI correctly on both occasions, rendering reliability and responsiveness testing inconclusive. The value of the PGI is significantly diminished by the fact that many people cannot complete it correctly. Future development of the instrument is appraised in the context of related measurement methods.