"If You Must Be Hospitalized, Television Is Not the Place"

Why do people think there are so many deaths in the ER? Perhaps because that’s how it looks on television. For a research paper called “If You Must Be Hospitalized, Television Is Not the Place,” the Israeli communications scholar Amir Hetsroni analyzed one season’s worth of the U.S. hospital dramas ER, Chicago Hope, and Grey’s Anatomy. It turned out that TV patients were drastically more likely to die in the hospital than real-life patients. They were also more likely to be young, seriously injured white men (preferably good-looking ones). Consider some details.

I ran into an old friend the other day whose actor husband is a regular on the TV show House. We caught up on friends and family, etc., including a few mutual acquaintances who have died since we last spoke. As we parted, I couldn’t help but laugh: at least these unfortunate deaths, I thought, were nowhere near as numerous as those on the kind of TV show her husband appears on. The chart at right is from p. 80 of the illustrated SuperFreakonomics.


Um...do you watch "House"? While it's not clear what the Princeton-Plainsboro Teaching Hospital's Emergency Room death rate is, the death rate for those patients who are assigned to House is close to 0%, since he tends to always figure out the problem at the last minute.

No, I wouldn't want to go to the other TV hospitals, but when it comes to House, sign me up! Not only can we likely get lots of cool pain killers from Dr. House (as in "Holmes"--get it?), but we also can enjoy the witty banter and partake in stinging existentialist dialogue.

What's not to love?

Imagine an episode of ER where the patient just has a cold.

That's because they are just playing doctors on TV.

Ian Kemmish

Maybe you're just looking at a skewed sample. Dr Finlay's Casebook ran forever in the 1960's, but I can't imagine anyone ever died in it.

Skewed samples are not unknown in real world hospitals, either. A regular source of tabloid "scoops" here in the UK is the observation that our "best" hospitals have some of the highest mortality rates. Tabloid journalists of course can't understand that this is because the undertake the most ambitious procedures and get referred the most unpromising cases.

Since most TV hospitals are located in large conurbations and employ superstar surgeons, it seems entirely reasonable that they'll see more fatalities, especially linked to violence, than small rural maternity hospitals....

Cash McDollar

"With my medical degree from Harvard, by my calculations. (Sob) I am deeply sorry, but you have ONLY two weeks to live." DUN- DUN-DUN.

"But if you get a unothordox, uncoventional, and frankly insane operation from the Young Handsome Doctor Boufant, you MIGHT get a one-in-a-million chance of beating it. But he may be a Double Agent Terrorist."

"But the world needs you to live... to prevent Global Thermonuclear War Armageddon since you are expert in diffusing nuclear polynomial codes. And since you are Supermodel, Rock Star, Race Car Driver scheduled for the pole position at Indy....And since you are also the President of the United States."


Don L

I think what we have here is a case of sample bias. These shows are classified as Hospital Dramas, not Hospital Documentaries. Even if they were based on real hospitals, the fact is that some episodes in a doctor's life are boring - you can only show so many births before you jump the shark. Thus, because the shows themselves are dramas, they tend to show the hospital incidents that have the most drama surrounding them, which tend to be those in which the chance of death is very real. The data you're citing tells us less about the dangers of fictional hospitals and more about which ailments the US population finds compelling (although I'll acknowledge that TV shows can be off regarding fear of rare conditions, like Mad Cow or Shark Attacks).

Eileen Wyatt

So does this mean if I work for a TV station in Minneapolis, it's statistically unlikely that my colleagues will be a collection of wacky individuals with a perpetual stream of comic life problems?

Are you now going to tell me that Dickens doesn't accurately capture the mortality rate and causes of Victorian England?

Story-telling tends to focus on the most dramatic incidents and skip the routine bits. Since one of Piaget's child development stages involves telling the difference between fiction and reality, I think TV audiences are expected to know that what we see in a fictional series isn't tested for statistical accuracy.

doc maybe

Someone explain why fertility problems is grouped with live births and infant diseases. Perhaps because all fall into the ob/gyn classification and can justify higher insurance rates?

Eric M. Jones

And if you have to fly someplace, don't take a TV airplane flight. Most commercial pilots will not see an engine failure or a crash in their working life.


I would love to know how "alcoholism and drug use" were difined to represent only 0.5% of US hospital admission.

My experience is that admissions related to drug and alcohol use are considerably more frequent. I have seen data suggesting that almost a third of all hospital admissions in Canada are alcolhol or drug related.

Oh, and good luck finding a mental condition to be present in only 2.5% of admissions...


Very true sir. In all my days visiting the hospital, I have yet to see a single death. It's quite regular on TV however.


Mike McGuire

On TV, police officers continually have their guns drawn and are continually being shot at. In real life, the average police officer will never draw his or her gun in their entire career. I can't address the frequency of their being shot at without their drawing their guns, but somehow I suspect it's much less often than is portrayed on TV. Statistically, I read once, sanitation workers are the real-world municipal workers most likely to die on the job, despite the real dangers firefighters and police officers sometimes face.

Following Piaget, it's probably best if we don't rely on legal shows to learn about the law, police shows to learn about police work or medical shows to learn about health care, or taking care of ourselves, for that matter.

Siddharth Surana

If I may say.. you guys have re-kindled my love for economics in the freak way. It's entertaining as well as informative.

Look forward to version 3, StraosphericFreakonomics!



This is data on deaths, not admissions.


You can also analyze NYC cop shows and compare them to real life. I believe that, the total deaths shown on L&O, L&O:SVU, L&O:CI and CSI:NY would be a significant fraction of actual NYC homicides (i.e., 24 shows per season, at least one death per show, 4 shows). And note that these would be the glossy murders, not the grubby ones that happen in Brownsville, etc.


It looks like we can greatly reduce health care costs and lower the hospital death rate dramatically if we stop trying to make babies and stop actually having babies. Babies appear to be much of the problem.

Too bad about that survival of the species stuff, it's costly.


The British Medical Journal published an analysis a while ago showing that living in a British soap opera was riskier than being a bomb disposal expert.