Reply to Hudorovic et al. Enhancing the transferability of region-specific findings: characteristics of Japan's cardiovascular surgery system

Abstract

Hospital volume is sometimes used as a surrogate measure for numerous care processes that directly influence patient outcomes. This is because individual processes and inherent technical skills are not well characterised and are difficult to measure. For some procedures, higher volume, particularly at the surgeon level, may translate directly to better clinical judgement and technical proficiency in the operating room. Such processes may be hard to transfer to lower-volume providers. Higher-volume centres may also differ with regard to other processes of preoperative and postoperative care, which could be adopted by lower-volume centres [1]. To better understand the volume—outcome relationship in cardiovascular surgery, our current study focussed on practices, including processes and structures, that could be related to perioperative mortality risk associated with any cardiovascular procedure. Processes refer to the care that patients actually receive, and structures are the environment in which the care is delivered [2]. We confirmed a volume—outcome relationship similar to that reported in the European Journal of Cardio-thoracic Surgery [3]. It must be pointed out, however, that health-care systems relating to cardiovascular surgery are not necessarily identical across regions. The relative strengths of processes and structures (especially the surgeon’s procedure volume and hospital procedure volume) might differ from one region to another. In this respect, we thought the ability to generalise our findings is limited, as Hudorovic suggested [3]; it would be useful to take steps to enhance transferability of our results. For readers from other regions, it might be useful to describe region-specific characteristics of the cardiovascular surgery system in Japan. Our previous study [4] indicated that an average of 3.3 cardiovascular surgeons participate in each cardiovascular surgery in Japan. This is because Japan does not have a nurse practitioner system or a physician assistant system. Even when a young surgeon leads the surgery, expert surgeons often participate as assistants. In additional, 95% of centres have meetings among cardiovascular surgeons more than once a week. Further, a majority of Japanese centres share information and experiences regarding all cardiovascular surgeries performed at their centres, thus the volume effect specific to a surgeon might be difficult to differentiate from the hospital volume effect. These region-specific characteristics might lead to differences in the relative impact of surgeon volume and hospital volume between Japan and other countries. If we could take such region-specific characteristics into consideration, even findings specific to a given region might become transferable (even if they cannot be generalised to all regions).