In many parts of the world summer is well on the way, but that means brand new doctors are on the way too. In July hundreds of new doctors start their jobs in hospitals in the United States, while hundreds of other doctors take on more responsibility after promotion. Something similar happens in Britain during the first week of August. The fact that a mass of doctors are all new to their roles, at whatever level, at the same time, has given rise to hospital horror stories.
Rumour has it that if you want to survive, you should avoid hospitals during those doctors’ first week at all costs. In the UK medical drama Cardiac Arrest the staff called it “the killing season”. Others refer to the day the new doctors start as Black Wednesday. In the US it’s known as the July Effect. But how much data is there to back up this folklore?
Surely new doctors are more likely to make mistakes? (Credit: Getty Images)
There are some studies which, at first sight, seem to back up the claim that these months are risky. In the US, for example, the National Bureau of Economic Research took data from 700 hospitals between 1993 and 2001. They found that mortality rates at major teaching hospitals rose by an average of 4% in July and August. That means the average teaching hospital saw between eight and 13 more deaths than usual. Patients also ended up staying longer in hospital. Over the next six months of the year, things gradually got better.
It sounds clear-cut. Surely the new doctors made more mistakes that led to patient deaths and then as they became more experienced the death rates went down. But does this mean that all hospitals are more risky at this time? Not at all. The risk only rose in teaching hospitals and even then in the busiest teaching hospitals there was no increase in deaths at all. This shows that if doctors are well-supervised and get the support they need, patient care doesn’t need to get any worse.
But not every study comes up with the same conclusion. Other studies covered every patient attending the emergency room at two Californian hospitals over a five-year period. There was no seasonal difference in summer death rates, but in July and August patients were more likely to suffer from preventable complications than those admitted in May and June.
More routine cases are often delayed, so the doctors could be left with the more serious semi-urgent cases
Then there are studies looking at particular procedures. Four thousand patients had emergency surgery for acute appendicitis in two American teaching hospitals between 1998 and 2007. They fared no worse in July and August than in any other month. Another Canadian study looked instead at children having surgery to insert tubes into their spines to divert cerebrospinal fluid. Here there was an advantage, although only a slight one, in going to hospital at a time of year other than July and August because the tubes lasted longer before needing to be replaced and there were fewer complications. This might suggest that better supervision is needed during procedures such as this when the doctors are new to the job.
Several other studies found that July and August was no more risky a time to be in hospital than at any other time year. But there was one study which was more worrying. It concentrated on surgery in teaching hospitals and they did find a lower survival rate in July and August than during the rest of the year. Some elements of the study, however, made it hard to know whether the extra deaths were down to the trainee doctors. The study made sure there was the same amount of emergency surgery included in each time frame.
But they couldn’t control for the fact that because a lot of patients and doctors are on holiday in July and August, more routine cases are often delayed, so the doctors could be left with the more serious semi-urgent cases during these months, and they are the kind of cases more likely to experience complications.
Winter death rise
They did collect some interesting data on the length of time it took for staff to prepare a patient for surgery and found that in July and August operations took an average eight minutes longer from the time of the first incision to the time of putting the dressing on at the end of the surgery. This isn’t dangerous, but might indicate that they were struggling more with the procedure. Or could it just indicate that they were new to the job and were taking the time to make sure it was done properly?
In this study the numbers of deaths did rise again in winter, which is commonly seen. There are various reasons for this. Blood pressure can be higher when it’s cold and there are more viruses and bacteria spreading around the community. This can be enough to lower survival rates in hospital.
The wrong doses of medicine may be behind the extra deaths recorded (Credit: Getty Images)
But what’s interesting about this study was that December saw a bigger spike in deaths than January and February, which is unusual. Could the reason be that more doctors take time off because of Christmas, leaving systems running less smoothly than they might at other times of year? If a disruption to the rota is enough to lower survival rates, perhaps that is what is happening in summer too.
This brings us to the question of exactly why a few studies have found that patients are more likely to die in hospital in July and August (remembering that others didn’t find that). Your first thought might be that more junior staff botch surgery or miss diagnoses. But an examination of more than 25 years of death certificates suggests a more straightforward reason: giving the wrong medication or the wrong dose of the right medication.
Survival rates are so high in hospital these days, that the extra risk is very small
The researchers found that in teaching hospitals there were 10% more deaths in July due to prescription errors than at other times of year. We can’t know whether it was new doctors who made the mistakes. And since this study began in 1979 it is difficult to extrapolate from so long ago, since training and hospital procedures have changed so much.
Too much coincidence?
In 2009, a team at Imperial College in London looked to see how many people died after being admitted to emergency departments in 175 British hospitals between 2000 and 2008, the Wednesday that the new doctors began their jobs and compared it with the previous Wednesday. At first sight there was no statistically significant difference in the number of deaths on the two days, but once they took into account age, gender and socioeconomic status, there were 6% more deaths on the day the trainee doctors start.
Again this study doesn’t tell us exactly why. Even the authors stress that you can’t necessarily blame the difference on the new doctors. Perhaps people had more serious emergencies on the second day. You could argue that it seems to be too much of a coincidence that the week the new doctors start, more patients die, but even then the difference is very small, so the chances of it affecting you are very slight indeed. We don’t know how they died, so we can’t tell how many might have been prevented, had a more experienced doctor been looking after them.
We can feel slightly reassured that the difference in survival rates is not by any means found in all studies, nor in non-teaching hospitals and when there is a difference, survival rates are so high in hospital these days, that the extra risk is very small. It’s not quite the killing season. And good job too, since you can’t usually choose when you get ill.
All content within this column is provided for general information only, and should not be treated as a substitute for the medical advice of your own doctor or any other health care professional. The BBC is not responsible or liable for any diagnosis made by a user based on the content of this site. The BBC is not liable for the contents of any external internet sites listed, nor does it endorse any commercial product or service mentioned or advised on any of the sites. Always consult your own GP if you're in any way concerned about your health.