|
|
|
|
|
|
|
would expect, he found many points of tension and resistance among the clinicians filling out forms, the administrative needs of the information-gathering bureaucracy (in this case tracking occupational illnesses), and the feelings of those doing the coding. He states: |
|
|
|
 |
|
|
|
|
Occupational doctors' vocabularies are tied to the local histories of the workplace. Occupational physician's access to objects is mediated by the instruments and the terms that peopleemployers, employees, representatives of personnelthemselves use in the workplace . . . there is no guarantee that local vocabularies for identifying reality coincide with administrative nomenclatures, except in the rare cases where the regulatory language is put into use by the employees and the employers themselves. There is, therefore, a conflict between the attention that doctors give to local universes and the standardized administrative definition of pertinent objects for judgment. (Dodier 1994, 6) |
|
|
|
|
|
|
|
|
He goes on to say that sometimes the doctors treat the coding schemes as black boxes. Sometimes they argue with them, bringing in medical authority and expertise; and at other times the exigencies of time simply mean they code in an ad hoc, even arbitrary fashion. |
|
|
|
|
|
|
|
|
A not unreasonable response to the combined ambiguities of design and certification practice in the ICD would be to throw up one's hands and walk away from any sense of data quality or certainty about the meaning of the ICD statistics. How can one know who is dying of what or where? And yet from the point of view of very large organizations, information, and diversity, this would be to abandon as well a great deal of rich information. This information includes not only content, but also methodological information about the ways in which software and its attendant categories become "frozen policy." The question is extremely complex. Global-scale, highly integrated computing systems are currently being built and augmented, such as the Web, and every day they transmit vast amounts of information around global networks. These tools, and the situation described here, demand new conceptual approaches for understanding the nature of this infrastructure. Two major lessons have emerged from this examination of the ICD as a coordination device: |
|
|
|
|
|
|
|
|
1. It is unrealistic and counterproductive to try to destroy all uncertainty and ambiguity in these sorts of infrastructural tools. By their very nature, classification systems need appropriate degrees of both to workonly in a totally uniform world (within a given specialty) would |
|
|
|
|
|