RT Journal Article
SR Electronic
T1 Common bile duct injury during laparoscopic cholecystectomy in Ontario: Does ICD-9 coding indicate true incidence?
JF Canadian Medical Association Journal
JO CMAJ
FD Canadian Medical Association
SP 481
OP 485
VO 158
IS 4
A1 B. Taylor
YR 1998
UL http://www.cmaj.ca/content/158/4/481.abstract
AB BACKGROUND: Recent reports in the scientific and lay press have suggested that bile duct injuries during laparoscopic cholecystectomy are common in Ontario. The reports were based on administrative data collected by hospital medical records departments and the Canadian Institute for Health Information (CIHI). The current study involved a direct inspection of hospital records to determine if the CIHI data accurately captured the rate of clinically significant bile duct complications. METHODS: For the period 1991 to 1995, records of bile duct injuries after laparoscopic cholecystectomy were independently evaluated to clarify the clinical significance of the complications. Of 21 Ontario hospitals for which data on complications had been reported in the media, 18 provided detailed information on all patients reported to have suffered bile duct complications classified by the hospital as "major". In addition, each institution provided information on a random sample of one-sixth of the patients who had suffered complications classified as "minor". The reviewer then examined each relevant hospital chart to assess the grade and significance of the reported complications. RESULTS: All 24 bile duct injuries classified by the hospitals as"major" were confirmed as major (clinically relevant) injuries. Of the 80 bile duct complications classified by the hospitals as "minor", 76 (95%) were irrelevant to patient outcome. The discrepancy between data collected and reported frequency of injury lies in the use of nonspecific coding methods. INTERPRETATION: The rate of significant bile duct injuries cannot be inferred from nonspecific codes taken from the International Classification of Diseases, ninth revision, and presented in hospital discharge records. Therefore, such data must be interpreted with extreme caution.