We’ve all heard the rumours: on the first Wednesday of August, as junior doctors around the country begin their first day of work, there is a small spike in the number of deaths in hospitals.
Many of us newly-qualified doctors are terrified that these rumours might actually be true.
After all, we went into medicine because we wanted to help people. The idea that we might actually be doing patients harm understandably concerns us quite a bit.
As a junior doctor about to start my first day of work today, the thought that my incompetence and ill-preparedness could result in patients dying is quite horrifying. But is there any truth behind the so-called ‘black Wednesday’ effect? What does the evidence actually say?
Where did this rumour come from?
There have long been suspicions that the influx of new junior doctors each August leads to problems in patient care. Anyone who works in a hospital will understand that if you flood a medical facility with a load of new, unexperienced staff, all on the same day, a degree of chaos and confusion is inevitable.
However, it wasn’t until 2009 that the first real evidence appeared that ‘changeover day’ really does have a detrimental impact on patient care in the NHS.
A study by researchers at the Dr Foster Unit, Imperial College London found that if you are admitted to hospital on this fateful Wednesday, your chances of dying are 6% higher than if you had been admitted the previous Wednesday in July.
Compared with similar studies of its kind, this study was of reasonably good quality. It compared a huge number of medical admissions from 175 different hospitals around the country. Also, rather than just taking a snapshot of what had happened in the last year, It looked at an average of what had been happening every August over the last 8 years.
A 6% increase is a relatively small rise
The first thing to say is that although the spike in mortality rates was statistically significant, it was not a particularly large effect. 6% is a relatively small rise in numbers of deaths.
We know that the season of the year, for example, has a considerably larger effect on your chances of dying after admission to hospital – studies have shown that there are around 27% more deaths in Winter than in Summer.
That said, any increase in patient deaths, no matter how small should be taken very seriously, especially if these deaths are potentially preventable.
Correlation does not imply causation
Another important caveat is that correlation does not imply causation. Simply finding that mortality rates increase during the same week that junior doctors start their jobs, does not prove that the new junior doctors are actually causing the excess deaths.
Perhaps, for some reason, the patients admitted on the first Wednesday in August are slightly sicker on average than those admitted the previous Wednesday.
Another potential explanation for the increased death rates could be that there was a greater proportion of consultants and other experienced staff on holiday at the start of August than the previous week. Again, there is no way of proving whether it is the effect of staff on annual leave, new doctors on the ward or a combination of the two, based on this study.
Is this a ‘new doctor effect’ or a more general ‘changeover effect’?
It’s important to bear in mind that the first Wednesday of August is not only the day that all newly-qualified doctors start their first jobs, but also the day that all other junior doctors ‘rotate’ and start a new job in a new hospital.
Therefore, as well as the ‘new doctor effect’, there’s also a more general ‘changeover effect’. Hospital wards will be staffed with whole teams of doctors starting work in an unfamiliar environment. Every hospital has its own slightly different systems, there will be new IT software to navigate and lots of opportunities for small hiccups which together culminate in getting behind on important tasks for patient care.
So what does more recent data show? Does the ‘Black Wednesday’ effect still exist?
Since the Dr Foster study was published in 2009, huge efforts have been made to minimise the impacts of junior doctor changeover day.
There have been significant changes in medical school curricula. The entirety of our final year of medical school is now very much focused on transitioning from knowledgeable but inexperienced medical students to competent and safe hospital doctors.
The prescribing of medications was highlighted as an area where newly-qualified doctors were making mistakes and our training now includes an additional exam and lots more focus in our training to ensure we don’t make errors around prescribing decisions.
Perhaps most importantly, we now do extended shadowing periods following our final medical school examinations, where our focus is solely on preparing ourselves for that very first day of work, rather than passing exams.
Other initiatives include ensuring adequate senior doctor cover during the changeover periods, high quality departmental induction on all units and reducing elective (non-urgent) work so that there are more resources and support available for junior doctors and for the care of acutely unwell patients. (The 2013 Academy of Royal Colleges Recommendations for Safe Trainee Handover are a nice guide as to what sorts of things are being done to imporve patient safety during this period).
However, there don’t seem to have been any further national studies to assess whether all of this has had an impact on patient care during the August transition period.
In 2013 there was a much smaller study which did not shown any increase in death rates on or after the first Wednesday of August. However, this study looked at patient death rates in just a single hospital, rather than across the whole NHS (as the 2009 Dr Foster study did).
Another study in 2013, instead of measuring hospital mortality rates, attempted to measure junior doctor performance and workload. This study concluded that there was no evidence of new doctors completing tasks more slowly or having a greater workload.
In 2014, a very informative review paper attempted to summarise all the recent studies looking at (1) UK medical graduates’ preparedness for starting work and (2) the effectiveness of the various different schemes which have been implemented to improve junior doctor preparedness for work.
The collection of studies seem to show that junior doctors are generally well prepared for history taking, physical examinations, practical skills and understanding their own limitations but generally less well-prepared in terms of safe prescribing skills, clinical reasoning and diagnosis, emergency management, safety and error reporting, ethical and legal issues and understanding how the clinical environment works.
However, the data also suggests that measures such as extended shadowing periods and induction have made a difference and new doctors nowadays are more prepared for work than they would have been back in 2009.
These problems seem to occur in healthcare systems around the world (not just the NHS).
Finally, it is worth noting that these problems are not unique to the NHS.
There is a wealth of evidence that similar difficulties arise in other health care systems on the day that new junior doctors start work or that doctors in trianing rotate from one job to the next.
A review paper from the US which summarised 39 different studies found that mortality increases and the efficiency of care decreases aroudnm the time of junior doctor changeover. But again, this data is old and there’s lots of reason to belive that our junior doctors here in the NHS are better prepared for their first day of work than those in the US, because of the various changes mentioned above that we’ve implemented over the past few years.
Unfortunately there’s no up-to-date evidence to say whether we’ve successfully thwarted the ‘Black Wednesday’ effect, but there has been lots done to minimise the impacts since the initial 2009 report and there have been studies which show this has improved junior doctors preparedness for their first day of work.
My gut instinct is to say that this additional emphasis during medical school on safe prescribing, the additional shaddowing periods and the improved induction schemes have really made a difference.
I’ve certainly had a fantastic induction period here at the West Suffolk Hospital. I’ve been made to feel about as familiar with the IT systems as can reasonably be expected and simulated training sessions on how to manage acutely deteriorating patients have made me feel a lot more confident.
I also know that there will be additional support from consultants, the ‘critical care outreach team’, pharmacists and the nursing team – the hospital has gone out of its way to make sure that staffing is optimised as we start work.
Essentially, everyone in the hospital is aware that this is a difficult time for us, that we will need a lot of help, and they’re looking out for us!
Of course us new doctors are all hugely stressed and nervous as we approach our first day of work. It’s only natural to feel wholly under-prepared, given that we’re taking on such a monumental responsibility for the first time.
That feeling of inadequacy makes it tempting for us to us allow the myth of ‘Black Wednesday’ to be propagated.
However, I’d encourage my fellow junior doctors to be positive and fight back against anyone who suggests that our patients may be less safe because of our presence on the wards today.
Remind the doubters that we’ve been to some of the best medical schools in the world, had several years of world-class preparation leading up to today and that, most-importantly, we’re all being supported by fantastic teams!
Good luck everyone!