In contrast, mediastinal widening on chest x-ray or computed tomographic scan or Gram stain of cerebrospinal or pleural fluid should lead an alert and knowledgeable physician to consider the diagnosis of anthrax, even though these tests may not be conducted until relatively late in the clinical course.
Incomplete observation sets were more likely to contain observations which should have led to an alert than complete observation sets (15.1% vs 7.6%, p<0.001), but less likely to have an alerting score correctly calculated (38.8% vs 30.0%, p<0.001).
This led to an alert being sent out over the embassy's Facebook page, resulting in 150 protesters assembling and maintaining a vigil.
Clusters of cases of endophthalmitis traced back to such contamination led to an alert being issued by the US Food and Drug Administration [ 71].
A total of 33.1% (515/1554) of the observation sets that should have led to an alert did not have an alerting score recorded.
Observation sets that should not have led to an alert but which had an alerting aggregate score recorded were more common in incomplete than complete observation sets (3.2% (118/3716) vs 0.9% (116/13 079), p<0.001).
Incomplete data sets were more likely than complete data sets to contain observations that should have led to an alert (15.1% (560/3716) vs 7.6% (994/13 079), p<0.001), but less likely to have an appropriate alerting aggregate score recorded (61.3% (343/560) vs 70.0% (696/994), p<0.001).
Table 1 shows the breakdown of recorded scores based on whether they would lead to an alert or not, compared to the expected action (alert or no alert) for correctly calculated scores based on the same vital signs data.
Incomplete observation sets should be avoided, as these are more likely to contain vital signs that should lead to an alert and more likely to be associated with errors that cause an alert not to be generated.
As incomplete observation sets were also particularly likely to contain vital signs that should have led to an alerting aggregate score, errors in incomplete observation sets are particularly likely to affect patient care.
