Article Text


Accuracy and completeness of the documentation of blood culture results
  1. J R Greig
  1. Derriford Hospital, Derriford Road, Plymouth, Devon PL6 8DH, UK

    Statistics from

    Heard and colleagues1 raise the thorny issue of how important microbiological results and advice should be recorded in patients’ notes and by whom. Reported here is a prospective audit identifying how accurately this is done. It was decided that the audit should concentrate on blood cultures yielding a clinically significant isolate, a finding that all members of staff should consider as important and worthy of prompt documentation.

    This work was performed in a 1045 bed hospital where all Gram stain and culture results from clinically significant blood cultures are telephoned by a medical microbiologist to either the attending doctor or to qualified nursing staff on the ward. It is expected that the information and advice given over the telephone should be promptly documented in the patient’s notes and that nursing staff would contact the patient’s attending doctor. Where it is clear from nursing staff that the patient is still septic, not responding to empirical treatment, on inappropriate treatment, or when some other medical action is necessary, the attending doctor is contacted directly. Because of time constraints personal visits to the ward are the exception rather than the rule. In a minority of cases culture results are imparted to attending doctors face to face. Of the 61 blood cultures reviewed, four were ultimately considered as contaminants. The microbiology departmental clinical database was used to identify what information was imparted, to whom, and whether this occurred on a weekday or at the weekend. The patient’s notes were scrutinised more than 24 hours after contact was made. The required minimum entry in the notes was correct Gram or culture result. The accuracy of this and any other recorded information was noted. It was felt that if simple microbiological information concerning clinically significant pathogenic infections was not being documented then more complex clinical information was unlikely to have been recorded any more faithfully.

    Of the 61 episodes reviewed, 16 (26%) were not recorded in the patient’s notes at all. One hundred and eleven individual contacts were made and 71 (64%) of these contacts were recorded in the patient’s notes. Not all of the 61 episodes necessarily had two contacts made, on 11 occasions the microbiologist wrote directly in the notes. It was of note that 60 of the 76 calls to doctors were documented in the patient record, whereas only 10 of the 34 calls to nurses were documented. There were marginally fewer documented calls at weekends (56% v 67% during weekdays) but this was largely because more weekend calls were made to nurses. In one instance it was unclear to who the telephone call was made to.

    The number of contacts documented is similar to previous studies that have looked at the accuracy of both telephone and face to face contacts. In similar studies 83% and 79% of clinically significant blood cultures had at least one entry in the patient’s notes.2,3 There seemed to be little difference if the result was reported over the telephone or face to face.3 An audit looking at day one results only reported a documentation rate of 54% but contaminated blood cultures, which may be less likely to be documented, were also considered.4

    This audit shows that advice given by microbiologists to other members of staff is recorded in the notes on fewer than two thirds of occasions, although where recorded it is generally accurate. The finding that over one quarter of clinically significant bacteraemias where not noted at all is troubling. This audit was not designed to measure whether these omissions led to adverse outcomes, although one might expect this to be the case on occasion. At Derriford Hospital, computerised reports of all culture positive are issued. There is generally a lag of at least 48 hours between initial culture and the issue of the report and probably a further delay before the attending doctor looks at the report. In many cases, this leaves a window when the patient’s condition is at its most critical and where important microbiological advice is not readily available. These data support the need for microbiologists to review all clinically significant bacteraemias and write in the patient notes. With all the other calls on the time of microbiologists, few departments can provide such a service. Without innovations such as electronic patient records or real increases in the numbers of medical microbiologists, the laudable advice of Heard and colleagues that “microbiologists should make relevant notes in patients’ records” is unlikely to be a realistic option in many hospitals in the UK.1


    View Abstract

    Request permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.