Think Twice About Operating on Fridays and Weekends and Stick to Golf Instead
I recall participating many elective major procedures on Friday nights and Saturday mornings during my residency training, thinking to myself that not only should I be home, but this just can’t be good for the patient…can it? Well, apparently not.
A new population-based study by Aylin et al. published in the British Medical Journal suggests that patients undergoing surgery on Fridays and weekends have significantly higher of both 2-day perioperative mortality as well as 30-day mortality. Utilizing the robust information provided by the English National Health Service (NHS), the authors analyzed over 4 million elective cases performed in England from 2008-2011 and found a crude mortality rate of 6.7 per 1000 cases. While overall mortality seems low, after adjusting for confounding variables the authors found a stunning 44% and 82% statistically significant increase in 30-day mortality if an elective procedure was performed on a Friday or weekend compared to Monday, respectively. When analyzing 2-day mortality, the authors found a whopping 167% increase in mortality on a weekend compared with Monday.
A “weekend effect” has been proposed in prior studies, however these studies for the most part analyzed emergency admissions and included emergency surgeries on patients that were likely to be much sicker than the average patient. What makes this paper different, and thus more significant, is that it only analyzed elective procedures and is the first paper to suggest that with each successive weekday, patients are at increased risk of mortality, culminating with the highest risk on Fridays.
Data on urologic cases within this study remain unknown, as urologic procedures were not selected for sub-analysis. However, overall analysis included all elective procedures, which must have included high-risk urologic procedures such as cystectomy, nephrectomy, partial nephrectomy, prostatectomy, RPLND, and endourologic procedures on infected stones. Therefore, this data should still have relevance for urologists performing such high-risk procedures.
Why is this happening? We know that major complications from elective surgeries happen within the first 48 hours postoperatively (Cavaliere F, et al.). Therefore, patients that have surgery on Friday or over the weekend are at their most vulnerable when the hospital is most short staffed. Additionally, there has been concern that the more junior faculty and trainees bear the majority of weekend coverage, and are therefore most often the primary points of care over weekends, leading to potential failure to rescue due to inexperience. Finally, there is the issue of cross coverage and dialogue between hospital staff during the week and the weekends. How much can a covering physician truly learn about a potentially complicated patient from a simple sign-out?
More importantly, what can we do? Ideally, major cases should be scheduled earlier in the week to allow the patients to have care while all hospital staff are available during the remaining week or so of recovery. Endoscopic and same-day procedures should be scheduled later in the week. However, is this realistically possible? OR time can often come at a premium and is difficult to come by in some busy hospitals, especially for junior faculty. Therefore, such a change would have to come from the top hospital administrators and likely would meet resistance from more senior faculty.
When asked by The Guardian regarding these results, Sir Bruce Keogh, cardiac surgeon and director of the NHS, downplayed the results, stating that when he performed open heart surgeries he would often intentionally operate on patients later in the week to get more time in the ICU over the weekend. With all due respect to Sir Keogh, I just do not see the logic in this approach, and feel we should take these results more seriously rather than downplay them. The data presented by Aylin et al. seems pretty convincing to me: while overall mortality is low, patients getting surgery later in the week and on weekends are getting inferior care leading to inferior outcomes. We need to acknowledge this data, not ignore it or diminish it, and come up with some kind of reasonable and fair solution to the problem.
What say you, Urology community? If any field can come up with a solution, it’s us. Somehow, we need a system that allows all surgeons, young and old, to perform higher risk surgeries earlier in the week to prevent potential complications happening under the watch of an undermanned, inexperienced hospital staff over the weekend. In the meantime, I will try to use my free weekends for spending time with my wife, golf, and watching sports while trying my hardest to perform major surgery earlier in the week. Not only will this please my wife, it will likely improve the care of my patients.
Keith J. Kowalczyk, MD
Department of Urology, Medstar Georgetown University Hospital, Washington, DC, USA. @KeithKow
Comments on this blog are now closed.
Great blog Keith! I used to do all day majors on a Friday and never liked it. I got home late Fri night and was in there most Saturday mornings. Not great for family life.
However the more serious issue is what this BMJ paper is suggesting – that there is risk to patients in doing major cases later in the week and especially Friday. For urology, I would argue that in experienced hands it is safer to do a bunch of robotic prostatectomies on a Friday than to do an afternoon list of TURPs. RARP complication rates are low in this setting and almost all go home the next morning as per care plan. Having the non-urology weekend team looking after the TURPs inevitably leads to more calls about haematuria and blocked catheters.
I now chose to do neither. Academic and consulting work Fri am, a long lunch then school pick up most weeks.
Great blog Keith. If only I could play golf on fridays. Have been thinking about this over the last few days and certainly has been big news in the press in the UK. Of course they were using relative statistics so loved by authors to maximise impact when actually the absolute values are very low. So I read the paper desperate to find all the possible flaws:
1) Case selection
2) Statistical manipulation
3) Inaccurate baseline data (NHS HES data is notoriously inaccurate)
4) Exclusions (patients with incomplete data were excluded – is this more likely for coding towards the end of the week with data not being completed properly)
5) Could semi-emergency patients be being mis-coded and are getting done at the end of the week before the weekend ?
… BUT 4,000,000 elective operations ! 27,000 deaths !
And then I get this uneasy feeling that I could probably believe it. NOT in my hospital of course! – but in all the peripheral hospitals I worked in as a trainee. Limited cover over the weekend. More junior doctors making decisions about patients they don’t know.
But does this apply to urology ? Impossible to know and no urological operations are specifically listed in the paper. It could well be true though for majors and minors … The TURP who doesn’t have his catheter removed and mobilise because the inexperienced doctor wasn’t sure about the colour of the urine … The cystectomy who is developing ileus but is still fed and aspirates … The missed TED stockings or LMWH.
But we must remember the differences are actually very small and patients must be reassured that elective operating is still very safe. I wish I could take the approach of some on Twitter:
@declangmurphy: Why do majors on fri? I do academic work and consult then long lunch. Better for all
@loebstacy: Agree, Monday and Friday are ideal academic days!
@daviesbj: the king operates M/T/W/R and rests Friday.
Sadly I can’t.
Great summary Keith. This has been an interesting topic of discussion on Twitter. Monday operations have a 0.55% mortality rate and Friday operations 0.82%. The extra risk is 0.27% of operations. This does not sound so bad but when we look at the real figures as in Table 2 of the paper, we can see that the mortality rate per 1000 admissions for elective surgical procedures is 5.5 and 8.2 for Monday and Friday respectively. This is an absolute difference of 2.7 deaths per 1000 admissions. That means that for a Friday elective operation, an additional patient dies for every 370 operations performed on a Friday instead of a Monday. I find it difficult to understand why these figures should be downplayed other than to protect the image of the NHS and to quell public anxiety. I am concerned that this effect will be evident in Australia as well and this is something that we need to look into as well.
The real test is whether anything is going to be done about these findings – we should also ask ourselves if we would have elective surgery on a Friday if we knew that we had a 1 chance in 370 of increased chance of dying compared to having an elective procedure performed on a Monday?
This is an interesting study and raises many important concerns about patients’ safety. If the various health care systems were robust enough to allow the flexibility, then most surgeons would prefer to avoid operating on Fridays/Saturdays. The National Health Service (NHS) in UK is a huge organisation with many resource constraints and surgeons have to adopt to the limitations imposed by the infrastructure.
It is a strange paradox that NHS keeps on expanding the pool of surgeons to deal with ever expanding work load but what has not changed over the years are the job plans. Most surgeons spend about 25% of their working week operating and the rest doing other activities. As a result there is a chronic problem of waiting lists. In order to deal with the waiting lists and comply with the government set targets, patients who are about to breach are scheduled for operations over the weekends.
Majority of the newly appointed surgeons/urologists are allocated Friday lists which are least popular because of safety concerns and social factors alluded to. This is the phase in consultants career when they are most likely to need help in complex procedures and are also most vulnerable to make mistakes besides the inconvenience.
However, infrastructure at present is not robust to allow the flexibility proposed. Hence the younger surgeons will continue to face the challenge of operating later in the week. However, if they aspire to do complex surgical procedures it is mandatory that patients’ are provided optimal post-operative care ideally by themselves or by one of the consultant colleagues rather than being left at the mercy of junior doctors.
Great article and well written. You summarise the important issues and problems nicely. However, there are those of us who have (public) operating lists on Fridays, and other days, but are so swamped with cancer work that we have no option other than to populate most or all Friday lists with complex surgery. We are flat-out trying to keep the work at bay. No majors on a Friday = longer waits = poorer outcomes. And from a health service perspective, inevitable exceeded waiting times = very bad trouble for them (and us, or so we are told).
The problem is not the day, but the provision of cover at weekends, as you allude to. We try to maximise safety by supporting this cover at weekends, which means every weekend is on call for the public system.
If too much is made of this topic before it is carefully thought through and solutions put in place, those of us at the sharp end who have no sensible choice but to operate on majors on Fridays, will be the first up against the wall when things go wobbly.
Will we be criticised for operating on Friday when we have a complications? Will patients want to know why, in the light of this evidence, we operated on them on a Friday if there is a complication? We know the hospital structure will not support the surgeons’ case when complaints come, even if we can demonstrate our access swipe cards were used at 0800 every Saturday morning of the year.
Thanks for doing this so promptly and well Keith.
There has been lots of discussion about this from urologists.
If you read the paper then the people who need to be talking about it are the GI surgeons and thoracic surgeons and I haven’t seen much from them. As someone married to a general surgical trainee we both feel it is really dangerous. One FY1, CT1 and SpR are meant to see 60 wards pts, 20 emergency pts, 5 ITU pts, ward referrals and then the day before elective pts while traipsing all round the hospital- Well Surprise Surprise, people with complications are missed! and how often do the general surgical consultants come and see their elective pts- NOT much!
This should serve as a big wake up call to how we do things. If you don’t feel appropriate care is available after an operation and you cant personally offer it- then you shouldn’t do it.
I think we have probably come up with most of the reasons this occurs
1. Lower staffing at weekends- lets be frank there are 20% or less juniors around at the weekend. They are focussed on the on call pts with pressing emergencies, they may not review elective cases from the day before until late pm.
2. Weekend staff are from other teams, not familiar with the operations and how pts are meant to be doing, or even from other teams cross covering
If there is a major emergency case then these people may get no review at all
3. Limited Consultant cover unless the operating surgeon makes the effort.
4. Bloods take longer to come back and other investigations much harder to get as radiology under same pressures.
5. Younger consultants allocated Friday lists.
I have a Friday list and have put a Sat am ward round into my job plan which I do EVERY Sat am at 0800 however I feel. I do like it on Friday doing RARP and RPN as no-one interferes with my list and i can be very focussed on the cases with little other distractions in the hospital.
Shamim’s points are excellent- younger surgeons need support and to operate on complex majors earlier in the week. We need more people with his dedication.
To start with if you operate on a Friday- come and see your pts and drive the process, if you cant then switch days.
What a great blog.
There’s no doubt this study has to be taken seriously and cannot be dismissed. The difficulty is knowing what to do with this information.
The possible interventions could be:
1) None – it says nothing about Urological Surgery or my institution (the others are rubbish)
2) Ban operating on fridays (I bags not being the one to tell our managers)
3) Running the hospital at full consultant capacity across all units for 7 days a week including elective operating on both days of the weekend (time to leave the country)
4) Devising a universal system for estimating operative risk and stratifying patients into early or late week procedures
There may be some merit in the final suggestion. As pointed out by Declan, it is not necessarily just the operation that determines risk – a difficult TURP in a patient who was on warfarin may be much more risky than a RARP in a fit 55yo man.
Things needing to be taken into account would be the patients comorbitities, level of current fitness relative to these comorbidities, the procedure and the relative technical difficulty of the procedure for this particular patient. Just the ASA and type of procedure would not tell the whole story. I would like to throw down the gauntlet to some of my more academic colleagues to come up with such a system we could all work with in light of these recent findings.
In relation to Shamim’s comments, I am very grateful being the most junior consultant on our unit and getting to do all my cystectomies done on a Monday and Tuesday!
Great points all.
Yes, it is administrative data, which has it’s flaws, but this is still an alarming trend, and regardless of exclusion criteria, it must reflect some truths. Where there is smoke, there is fire.
Agree that probably more complications following a TURP than RARP, but this is usually clot retention and not lethal. A PE would be more likely to happen after RARP than TURP.
One thing to consider: These findings conclude that the increased mortality over a weekend is due to inexperienced postoperative coverage. Well, as we have mentioned, junior faculty are also more likely to operate later in the week. Perhaps we are seeing differences in mortality not due to postoperative care, but due to surgeon inexperience. The results were not controlled for based on surgeon experience, which may have altered these results.
Regardless, I do think it makes much better sense. We need to see more real results, but just like the checklist mentality has been implemented in most ORs and hospitals, perhaps now a scheduling protocol for hospitals should be implemented in which surgeries are categorized as high, moderate, and low risk and can only be scheduled on certain days.
I know, we are living in a dream world thinking this can be accomplished easily, but we need to start somewhere!
Keith
PS – I rarely get to Golf on weekends either, but one can dream…my chairman certainly golfs on Fridays on occasion! One day….
I think that we should also think about “handoffs”. There are more handoffs over the weekends in most centers, and these create less and less informed clincians.
This was a hot topic on Twitter, and many of these ideas have been thrashed out already, but I think Matt Bultitude’s points are well made. You can argue that it’s not an RCT, there may be bias etc., but anyone who has worked as a clinician must have a gut feeling that it’s a real effect based on their experience. No-one wants to be in the hospital on the weekend, so the job falls to a skeleton staff of juniors who are typically overworked. For many of us going through this period, the goal was simply to survive until Monday. It’s harder to get other specialties to give advice, get imaging, some diagnostic tests, and there may be a degree of fear about having to call the boss. I believe that evidence has existed for some time that death rates in hospitals are higher across the board over weekends & holidays. I have not noticed any efforts to address this issue in any hospital I have worked in. This is a failure. We now have evidence that we could be (are) putting patients at risk by operating on them on certain days of the week. Why on Earth would we keep doing that? This study is ammunition for us to give this problem to theatre managers, and demand that they “manage” surgery so that majors aren’t done at the end of the week. Will it be difficult? Certainly. Will it require dramatic changes in culture? Undoubtedly. But if the alternative is to continue risking lives, our role is clear in driving this change.