Jump to comment:

• Response to article 'The impact of 24hr consultant shop floor presence on emergency department performance: a natural experiment' by Christmas, Johnson and Locker, EMJ May 2013, Volume 30, Issue 5, pages 360-362.
  Rod Milne
  Published on: 29 April 2016

• Published on: 29 April 2016

  Response to article 'The impact of 24hr consultant shop floor presence on emergency department performance: a natural experiment' by Christmas, Johnson and Locker, EMJ May 2013, Volume 30, Issue 5, pages 360-362.

  • Rod Milne, Consultant EM

  Emergency Department,

  North Tyneside General General Hosp[ital,

  Rake Lane,

  North Shields.

  Dear Editor,

  With reference to the article 'The impact of 24hr consultant shop floor presence on emergency department performance: a natural experiment', by Christmas, Johnson and Locker, EMJ May 2013 Volume 30, Issue 5, pages 360-362.

  Some initial comments arise from reading...

  Show More

  Emergency Department,

  North Tyneside General General Hosp[ital,

  Rake Lane,

  North Shields.

  Dear Editor,

  With reference to the article 'The impact of 24hr consultant shop floor presence on emergency department performance: a natural experiment', by Christmas, Johnson and Locker, EMJ May 2013 Volume 30, Issue 5, pages 360-362.

  Some initial comments arise from reading this article. Whilst, there is no significant difference in the mean number of patients, in the department, at the start of the night shift (table 1), there are slightly less on the consultant nights. The medical staffing (table 3) would appear to be slightly better on the consultant nights (for SHO, middle grade and other consultant.) Whilst using an age and presentation by ambulance, as surrogate markers of case mix, one wonders why the usual triage categories (P1-4) are not used. The figures in Table 2 are percentages for each category. One might surmise that on the consultant nights, which are predominantly Friday, Saturday, Sunday (69% versus 38% for middle grade nights) that the case mix would be very different to weekdays, in that there would be more minor cases in the age band 16-65, who may arrive by ambulance, but end up being fairly minor cases. Whilst, the percentages in each category may be very similar, this may not be reflected in the complexity of the actual cases. Thus, the median waiting time and length of stay may be less for the consultant nights, this may be due to a greater proportion of cases being of less severity or taking less time to sort out. The mean number of cases presenting per hour is roughly one patient per 2 hours more for the consultant nights. Equally, one could argue that the busy weekend nights (although not that much busier by the data provided), with more intoxicated patients, who take longer to manage, has been more efficiently managed on the consultant nights. However, it is difficult to tease out the true complexity of cases, to say with certainty that this is so. One might empirically say (and hope) that having consultants around makes the department run more efficiently, by offering timely advice and active management (perhaps just by having the 'boss' around), but the figures show that both groups met the 4 hour target and had no difference, in the surrogate marker of safe discharge, namely, proportion of patients returning within 7 days. One wonders why the authors have used median waiting and length of stay times (one assumes the spread is the median times for all nights), which again takes no reference to the case mix and how many cases were very minor and took little time to see and more complex cases which take longer to see. These factors will not be obvious when a median value is used. The proportion of cases admitted on consultant nights is slightly less compared to middle grade nights - could this be due to more minor cases presenting on weekend nights, who do not require admitting and thus accounting for a smaller percentage of cases, overall, requiring admission? Either way, a 3% reduction in admissions of say 30 patients per night, equates to 1 patient per night. This article states that there is no significant difference in any of the outcome measures in Table 5, comparing the daytime, when a consultant is not available to be around, as they are doing the nightshift, versus when they are available and the middle grade is doing the night shift ie. the additional consultant makes no difference, during the day. This is an intentionally provocative statement, as obviously, as has already been mentioned in the article, there are other consultant duties that are not measured. At first sight it looks as though having consultants working nights decreases the median waiting and length of stay times and reduces the proportion of patients admitted. However, I would propose that it is not as straight forward as this. We are not comparing like for like - compare consultants versus middle grades nights, purely on weekends and purely weeknights. The article states that some of the reduction in process times is due to consultants seeing patients more quickly. The College recommendations are that consultants will work in a supervisory role overnight, or are they to be turned in to F2's for the rest of their careers? These comments are deliberately provocative, but do show that figures can be used to show what you like and that they do not tell the whole story. Managers may read articles such as this and see headline of 'consultants reduce waiting times and reduce admissions' - well, they reduce the median wait on weekend nights in a group of patients that possibly includes more minor cases and in either case, all of them are seen within 4 hours and if your department admits 30 cases overnight, then you will reduce the admissions by 1 per night and your marker of safe discharge is no different.

  Conflict of Interest:

  None declared

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top