Frank Snoek, VU University Medical Center Amsterdam
13 February 2007
In her feedback on FDA's draft guidance on Patient Reported Outcome (PRO) measures, professor Bradley among other issues, underscores the importance of making a distinction between health status and quality of life. She cites her commentary in the Lancet (2001), where she gave the example of the influential United Kingdom Prospective Study (UKPDS). She also cites the editorial with Tamburini in this journal (2003) citing the Lancet commentary, using the UKPDS as an example of misuse of the term 'quality of life'. For those not familiar with diabetes literature, the UKPDS was a large prospective study in type 2 diabetes that found diabetes-related complications but not intensive therapy to have a negative impact on the Quality of Life of Type 2 diabetes patients (1999). According to professor Bradley the UKPDS used health status measures, and quality of life was not measured. She bluntly states that "this is a very different conclusion and a far less desirable one than the one reached erroneously by the UKPDS authors". It is unclear to me how anyone could know what the desired outcomes were for the UKPDS study group, unless the authors had made this public, which to my knowledge they never did. I leave it to the UKPDS authors to reply to professor Bradley's speculation. More importantly, professor Bradley fails to report that in the UKPDS a number of validated patient reported outcomes were used to determine the impact - if any - on the patients life quality, of diabetes complications, hypoglycaemic episodes and intensive therapy per se. Next to the EQ5-D, the Profile of Mood States (POMS), The Cognitive Failures Questionnaire, questions regarding symptoms, and work satisfaction were administered, tapping into different relevant domains of psychological and physical functioning. We can debate whether the participating patients should have been asked the question "How do you rate your quality of life", and how would you rate it if you did not have diabetes/intensive therapy/hypoglycaemic episodes/complications" but I seriously doubt that this would have provided us with additional information. Contrary to what professor Bradley suggests, the findings of the UKPDS in my view have added important knowledge to the field of quality of life research, showing that complications (and not diabetes or intensive treatment per se) reduce subjective well-being in type 2 diabetes, and that those who reported more frequent hypoglaemic episodes also reported more tension, overall mood disturbance and less work satisfaction.
We all struggle with the concept of "quality of life" and we should try to avoid abuse of the term. However, to repeatedly take the UKPDS as a bad example of mixing health status and "quality of life" is not justified and indeed misleading. Such comments are likely to cause more confusion rather than clarity.
To add to the discussion on perceptions of health as opposed to quality of life, it is of note that (self-reported) mood and self-rated health are among the strongest predictors of clinical outcomes and mortality (e.g. see Hennesy et al.,1994; Nicholson et al., 2006). In the context of medical care they seem appropriate and discriminant questions, based on people's perceptions of how well they feel they are doing.
How far off a persons' 'quality of life' can that be?
Terminology or context?
13 February 2007
In her feedback on FDA's draft guidance on Patient Reported Outcome (PRO) measures, professor Bradley among other issues, underscores the importance of making a distinction between health status and quality of life. She cites her commentary in the Lancet (2001), where she gave the example of the influential United Kingdom Prospective Study (UKPDS). She also cites the editorial with Tamburini in this journal (2003) citing the Lancet commentary, using the UKPDS as an example of misuse of the term 'quality of life'. For those not familiar with diabetes literature, the UKPDS was a large prospective study in type 2 diabetes that found diabetes-related complications but not intensive therapy to have a negative impact on the Quality of Life of Type 2 diabetes patients (1999). According to professor Bradley the UKPDS used health status measures, and quality of life was not measured. She bluntly states that "this is a very different conclusion and a far less desirable one than the one reached erroneously by the UKPDS authors". It is unclear to me how anyone could know what the desired outcomes were for the UKPDS study group, unless the authors had made this public, which to my knowledge they never did. I leave it to the UKPDS authors to reply to professor Bradley's speculation. More importantly, professor Bradley fails to report that in the UKPDS a number of validated patient reported outcomes were used to determine the impact - if any - on the patients life quality, of diabetes complications, hypoglycaemic episodes and intensive therapy per se. Next to the EQ5-D, the Profile of Mood States (POMS), The Cognitive Failures Questionnaire, questions regarding symptoms, and work satisfaction were administered, tapping into different relevant domains of psychological and physical functioning. We can debate whether the participating patients should have been asked the question "How do you rate your quality of life", and how would you rate it if you did not have diabetes/intensive therapy/hypoglycaemic episodes/complications" but I seriously doubt that this would have provided us with additional information. Contrary to what professor Bradley suggests, the findings of the UKPDS in my view have added important knowledge to the field of quality of life research, showing that complications (and not diabetes or intensive treatment per se) reduce subjective well-being in type 2 diabetes, and that those who reported more frequent hypoglaemic episodes also reported more tension, overall mood disturbance and less work satisfaction.
We all struggle with the concept of "quality of life" and we should try to avoid abuse of the term. However, to repeatedly take the UKPDS as a bad example of mixing health status and "quality of life" is not justified and indeed misleading. Such comments are likely to cause more confusion rather than clarity.
To add to the discussion on perceptions of health as opposed to quality of life, it is of note that (self-reported) mood and self-rated health are among the strongest predictors of clinical outcomes and mortality (e.g. see Hennesy et al.,1994; Nicholson et al., 2006). In the context of medical care they seem appropriate and discriminant questions, based on people's perceptions of how well they feel they are doing.
How far off a persons' 'quality of life' can that be?
Competing interests
No competing interests.