We thank Catalá-López and colleagues [1] for their response to our article recently published in Value in Health and for drawing our attention to their own recent work on the burden of disease and drugs authorized by the European Medicines Agency between 1995 and 2009 [2,3], published outside the time frame of our main literature search. In addition, they cite other articles focused on research funding and expenditure in relation to disease burden [4,5,6,7]; however, this was not the focus of our study.

As Catalá-López and colleagues state, many of the technologies we identified were coded as being indicated for “other” categories within disease groups (e.g., “other malignant neoplasms”). It is not, however, true to say that these represent unspecified indications; all technologies included in “other” categories had specific indications associated with them. For instance, “other malignant neoplasms” included rarer forms of cancer such as glioma, and “other digestive diseases” included Crohn's disease. As we state, technologies with nonspecific indications were included only in the second-level analyses (e.g., using broader headings such as all “malignant neoplasms”) but removed from the analysis at the third (individual disease) level. We acknowledged in our article that some categories would benefit from further subdivision; however, this does not imply that the disease categories were themselves ill-defined.

We do not therefore consider the alternative version of the figure presented to be an appropriate representation of our findings. Excluding conditions classified within the “other” categories omits a considerable proportion of the available data on individual technologies (representing real innovation). Our reported correlation for the third-level analysis was calculated by using 102 individual specific disease categories for the identified technologies, whereas the alternative figure uses just 18 of the highest ranking specific disease categories, creating an arbitrary data set that does not provide a complete picture. Indeed our conclusion that there is, at best, a weak association between innovation and burden of disease corroborates Catalá-López and colleagues' previous work [2] and agrees with their interpretation of our own study data.