For those of us with a secure job, this might come as good news. For those of us still flailing about in the temporary market, we might be a little worried. The delicate issue of 24 hour in-house staffing by fully qualified intensivists has steadily lifted from a quiet back corridor whisper to a genuine topic of conversation. It has been part of the “jobs for post Fellowship intensivists” concerns at the last two New Fellow’s Conferences, held in conjunction with the CICM Annual Scientific Meeting, and comes up at various CICM committee meetings. The major themes of clinical benefit to patients and the need to create positions for newly qualified intensivists appears to be pushing us inevitably towards a 24/7 model. |
However, all is not necessarily as it seems, as is so often the case in medicine. Wallace et al have published a retrospective cohort study (“Nighttime Intensivist Staffing and Mortality among Critically Ill Patients“) in this week’s NEJM in which they attempt to address the expectation that if having an intensivist involved in a patient’s daily management improves outcomes (Pronovost et al. JAMA 2002;288:2151-62), then having one available at night should be even better.
The study is a retrospective one and so it can really only suggest association, but this would be a tough topic to try and convert into a multi-centre RCT (Take, for example, the MERIT trial). However, it was fairly well conducted and the group did their best to try and cover for inherent deficits as best they could. Acknowledging all of the limitations, the end-point conclusions were:
- If you run a low staff intensity ICU (i.e. a specialist intensivist is not necessarily involved in a critically ill patients’ management [Remember, this is an American study where there are a mix of ICU models, including closed, open and mixed specialist care]), then in-hospital patient mortality appears to be improved by having in-house night coverage by an intensivist.
- If you run a high staff intensity ICU (i.e. where intensivist involvement in the care of a critically ill patient is mandatory, or the intensivist is the primary physician), then there is no in-hospital mortality benefit by having a intensivist on the floor over night.
- Interestingly, night time staffing with any form of intensive care physician, including resident (or junior medical officer / trainee), available on the floor, is associated with improved in-hospital mortality.
In summary, the results of this study suggest that if you have decent day time intensivist staffing of your ICU, providing in-house night time staffing by intensivists adds nothing to the in-hospital mortality outcome. Having night time staffing by junior medical officers with specialist intensivist oversight readily available, also improves the mortality outcome; though whether the effect is additive was not addressed. Largely, this is the model of care provided in most Australian public intensive care units, which generally run a closed unit model.
However, there are some additional considerations. The primary outcome measure was in-hospital mortality. The study did not assess ICU and hospital length of stay, duration of ventilation or nosocomial morbidity rates. Furthermore, the authors acknowledge that they defined night time intensivist staffing by title only, not by their overnight activity. In other words, the intensivist may have been physically present in the hospital, but we don’t know if they were up and about or fast asleep for the duration of the shift. Finally, this retrospective study collected its data by dredging the data that units sent to the authors upon request, so there is likely to be a good degree of selection bias, with strongly motivated units who may be providing a number of other high quality packages of care possibly over-represented.
The paper is certainly worth a read and likely to be a hot topic for debate and you can access it here.
The accompanying editorial is here.