Marlot Kuiper

56 Connective Routines together (also Timmermans & Berg, 1997) I purposely adopt the widely cited definition of Bowker and Star. This broad and neutral definition of standards fits the content of this dissertation best, since I argue that the crux is in different types of standards, since different types of standards differ in both what they aim to render uniform, and they ways in which they (can) transform professional work. To create an overview of all these different kinds of standards, Timmermans and Berg ought “to create some uniformity in the many entities that fall under the standards heading.” Somewhat ironically, this thus implies that they ought to create uniformity in that what [standards] strives for uniformity. They do so by distinguishing for ideal typical categories of standards; design standards, terminological standards, performance standards and procedural standards. Design standards strive for uniformity in systems, for example technical specifications that allow for a smooth transmission of medical and administrative data between systems, or even the size of hospital beds. Terminological standards try to ensure stability of meaning of concepts over different sites and times, for example regarding consistency of terms in Diagnose Related Groups (DRGs). In contradiction to design and terminological standards, procedural standards interfere with work practices by standardizing work processes. Whereas design and terminological standards determine the contours of professional work, procedural standards thus intervene with practices as such. Procedural standards herewith provide a window of opportunity to restructure how work is conducted. An example of a procedural standard is a risk screening protocol for decubitus or a Surgical Safety Checklist. The fourth category of standards consists of performance standards . By setting performance standards, such as the amount of surgeries to be performed, the rate of in-hospital mortalities, or adherence to guidelines, professional work becomes measurable and thus comparable (De Bruijn, 2002). In this dissertation, the focus is on what Timmermans and Berg (2003) defined as ‘procedural standards’. However, in table 2 I indicated a thin line between procedural standards and performance standards. Although on the one hand they can be distinguished as two separate categories, it is argued that procedural standards increasingly become performance standards (see e.g. Timmermans, 2005). If for example adherence to a risk screening protocol for decubitus is used by external actors for the sake of performance measurement or reimbursement, procedural standards thus actually become performance standards. This dynamic relationship between procedural standards and performance standards