File under correlation-is-not-causation-but...
Sometimes even correlations are too significant and important to just be fobbed off by epidemiological cliché. This collection contains articles repeatedly showing association between doctor characteristics, particularly gender, and patient outcome.
Although most recently shown by Wallis in JAMA Surgery (2021), gender-outcome associations are depressingly not new.
- Female patients treated by male surgeons more commonly experience post-operative complications and death than when treated by female surgeons. (Wallis 2021)
- Care from male surgeons and/or anaesthesiologists is associated with longer lengths of stay after cardiac surgery. (Sun 2021)
- Female heart-attack patients are less likely to survive when treated by a male physician than a female physician. (Greenwood 2018)
- Treatment from female surgeons is associated with a lower 30 day mortality than the same from male surgeons. (Wallis 2017)
- In-patient care from a female physician is associated with lower 30 day mortality and readmission rate among elderly patients. (Tsugawa 2017)
The cause of this gender outcome disparity is unclear, and importantly these studies are hypothesis forming, rather than proving. Nonetheless both Wallis (2021) and Greenwood (2018) hint at causes, namely a lack of experience treating female patients for some male doctors, and consequential lesser understanding of gender-disease differences.
The temptation when attempting to understand this is to descend into medical gender essentialism – ironically, probably a contributor to the actual outcome disparities.
A similar doctor-outcome disparity is seen with age. Among physicians, care from older doctors was associated with worse outcomes (Tsugawa 2017), yet for surgeons older age conferred better outcomes (Tsugawa 2018; Satkunasivam 2020). Causes here are possibly a nexus between experience, up-to-date knowledge and work volume – but also, still unclear.summary
There has been some observational evidence that a greater depth of anesthesia, as measured by BIS, may be associated with an increase in post-operative mortality. In particular the association of the "triple low state" (low BIS, low volatile-ET, low MAP) with post-operative mortality is worrying.
Completion of the Balanced Anaesthesia Study Group’s large RCT looking at this issue however brings us as close to a final word as we may expect. Short et al. (2019) showed no difference in 1-year mortality for older patients undergoing major surgery, whether they received a deep (BIS target 35) or light (BIS target 50) general anaesthetic.
It is likely that earlier observational studies were showing the consequences of intraoperative hypotension resulting from anaesthetic depth, rather than anaesthetic depth itself.summary