Speight et al. cited the Food and Drug Administration (FDA) 2006 draft guidelines on Patient Reported Outcome measures which advised against questions requiring judgements about hypothetical situations. While this advice is sound in some instances, there are major benefits of the design of ADDQoL items which ask respondents how aspects of their life would change if they did not have diabetes (see response to FDA draft guidance by Clare Bradley published at http://www.fda.gov/ohr ms/dockets/dockets/06d0044/06d-0044-EC18-Attach-1.pdf). A direct question about the impact of diabetes often elicits optimistic coping responses suggesting little or no impact, while asking ‘but what if you didn’t have diabetes?’ would elicit a more realistic response. Speight et al. also mentioned conflicting views about whether the impact of a condition on domains of life should be weighted by the importance of each domain to the individual’s QoL. On the one hand, they cited criticism of the DQOL which does not allow respondents to indicate applicability or relative importance of domains. On the other hand, they stated that individualized measures such as the ADDQoL have been criticized for weighting the impact on each applicable domain by the domain’s importance to the respondent’s QoL. Unfortunately, they provided no references to support this criticism. We have demonstrated the usefulness of weighting by importance using the ADDQoL and related measures (e.g. [2,3]). Even where average weighted and unweighted scores correlate highly, weighting considerably influences specific domain scores in clinically important ways. Weighting also changes the ranking of domains compared with ranking of unweighted impact scores. In the MacDQoL for people with macular disease, 23 / 26 domains changed places, with six moving three or more places. In the RetDQoL for people with diabetic retinopathy, 16 / 26 domains changed places, with nine moving three or more. The retinopathy sample had less visual impairment and a high proportion indicated no impact on several domains. If impact is rated as zero, weighted impact will always be zero; thus, where samples report more impact, weighting by importance will have more influence. Some measures of life satisfaction appear to be little affected by importance ratings [4] and our group has shown that the DTSQ does not require importance ratings, as respondents felt that all items measure important aspects of treatment [Singh H. Psychological Aspects of Diabetes Management in South Asian and White Men and Women With Diabetes. Unpublished PhD thesis. Royal Holloway, University of London, 2007]. However, these findings cannot be extrapolated to QoL measures such as the ADDQoL, where importance of domains varies markedly across individuals. Some of the information Speight et al. provided in Table 1 is outdated. We provide correct information here in Table 1. On page 323, Speight et al. directed the reader inappropriately to a 1988 publication in Diabetic Medicine instead of the original development of the DTSQ status version published in 1990. Development of the change version is published elsewhere [6]. Up-to-date lists of language / country versions available for questionnaires developed by our group can be found at http:// www.healthpsychologyresearch.com alongside information about access to questionnaires and relevant publications. L. S. Brose*, J. Mitchell* and C. Bradley† *Department of Psychology and †Health Psychology Research Ltd, Royal Holloway, University of London (RHUL), Surrey, UK Competing interests CB is Director and majority share holder of Health Psychology Research (HPR) Ltd which licences her questionnaires, including the DTSQ, W-BQ12 and ADDQoL, for others to use. LB’s PhD studentship at Royal Holloway, University of London is funded by HPR Ltd. References 1 Speight J, Reaney MD, Barnard KD. Not all roads lead to Rome—a review of quality of life measurement in adults with diabetes. Diabet Med 2009; 26: 315–327. 2 Mitchell J, Bradley C. Design of an individualised measure of the impact of macular disease on quality of life (the MacDQoL). Qual Life Res 2004; 13: 1163–1175. 3 Brose LS, Watkins J, Bradley C. Retinopathy-dependent quality of life questionnaire (RetDQoL): psychometric development. ISOQOL Research meeting abstracts. Qual Life Res supplement 2007; A-60, no. 1198. Available at http://www.isoqol.org/2007mtgabstracts.pdf Last accessed 11 August 2009. 4 Trauer T, Mackinnon A. Why are we weighting? The role of importance ratings in quality of life measurement. Qual Life Res 2001; 10: 579–585. 5 Howorka K, Pumprla J, Schlusche C, Wagner-Nosiska D, Schab- mann A, Bradley C. Dealing with ceiling baseline treatment satis- faction level in patients with diabetes under flexible, functional insulin treatment: assessment of improvements in treatment satis- faction with a new insulin analogue. Qual Life Res 2000; 9: 915–930. DOI:10.1111/j.1464-5491.2009.02853.x The use of hypothetical scenarios and importance weightings when measuring the impact of diabetes on quality of life. A response to Brose et al. Diabetic Medicine invited us to contribute a brief review and expert opinion to demystify the concept of ‘quality of life’ (QoL) and its measurement in diabetes. ‘Not all roads lead to Rome…’ [1] offered an overview and critique of ten measures frequently used to assess QoL in diabetes. After critically appraising the content of these measures at an item level, we conceptualized seven of these as more accurately assessing health status, treatment satisfaction or psychological well- being—related to but conceptually distinct from QoL. For the DIABETICMedicine Letters ª 2009 The Authors. Journal compilation ª 2009 Diabetes UK. Diabetic Medicine, 26, 1075–1079 1077