Surgical Safety Checklists May Not Improve Quality

Aaargh!!!  I get tired of having one after another of the wonderful ideas to improve health care quality run into the brick wall of research reality.  The latest casuality is the surgical quality checklist.     (NEJM Article)   Based on early research, surgical safety checklists have seen widespread implementation with the belief that they can reduce both complications and mortality related to surgeries.  This study was conducted in Ontario, Canada, which has mandated public reporting about rates of adherence to a checklist beginning in July 2010.  The researchers compared outcomes before mandatory use of the checklist and after implementation, for all surgeries at about 130 hospitals in the province.  The primary outcomes were  mortality in the hospital or 30 days after the surgery.  They also looked at length of stay, rates of readmission and ER visits after discharge.  The usual morbidity and socio-demographic adjustments were made.  After implementation, use of the checklists was very high, well over 90% of procedures were covered in most hospitals.

The results, however, were disappointing.  There was no significant decrease in mortality, readmissions, complications or ER visits.  Length of stay showed a very small decline.  There was a decline in unplanned returns to the operating room, but there were increases in deep vein thrombosis and ventilator use.  No individual hospital showed a significant decrease in mortality.  Six hospitals did have significantly fewer complications after introduction of the checklist but three had significantly increased complications.  The results show the value of repeated experiments or research on the same topic and the need to avoid common biases that can affect research results.  The authors posit several possible reasons for the lack of benefit.  One not covered may simply be that checklists actually make people less careful–they think following the checklist will be adequate and are actually less careful in some regards.