“Alcohol Helps Heart Bypass Patients!” – good news or bad reporting?

16 Dec

“Light alcohol consumption was associated with a 25% reduction in additional heart procedures, heart attacks or strokes in a study by Italian researchers, presented at the annual meeting of the American Heart Association in Chicago.”

This report was distributed around the world by Reuters*, the venerable U.K.-based news agency. Trouble is, after the health journalism watchdog Health News Review got finished with their assessment of this news reporting, they awarded Reuters a rare score of zero on their six-point quality scale.  For example, their pet peeves:  

  • using causal language in the headline – “Bypass patients can benefit from a few drinks” – for an observational study that cannot prove cause and effect
  • no discussion of the limitations of drawing conclusions from observational studies
  • not a word about the potential weaknesses in a study that relied on people to fill out a questionnaire about their alcohol consumption
  • using only relative figures – not absolute – to describe both benefits and harms.

Before we examine the six important reporting areas that Reuters failed so miserably at, learn to think like a Health News Reviewer yourself. For example, their site warns of these seven words that journalists should never use in writing about health news:

The words are:

  • Cure!
  • Miracle!
  • Breakthrough!
  • Promising!
  • Dramatic!
  • Hope!
  • Victim!

“Each is a vague – sometimes meaningless – term when used in a health care context. Granted, they are exciting terms that might help sell papers or move a reporter’s story onto page one or into the first news block, but they can be dangerous terms that mislead vulnerable people.”

So here’s how this Reuters story got that rare score of ZERO out of a possible six points from Health News Review:

1.  Evaluate the quality of evidence? – NOT SATISFACTORY

We’ve said before that news organizations that make causal claims about observational studies should have to write on the blackboard hundreds of times: “Association does not equal causation.”

So using phrases like “can benefit from a few drinks” in the headline is wrong.

There was no discussion of the limitations of drawing conclusions from observational studies – and not a word about the potential weaknesses in a study that relied on people to fill out a questionnaire about their alcohol consumption.

2.  Quantify the potential harms? – NOT SATISFACTORY

Most of the story focused on benefits. The story reported:

“Bypass patients with a condition called left ventricular dysfunction who were heavy drinkers, defined as having more than six drinks daily, were twice as likely to die from heart problems.”

But twice as likely as what?  Twice as likely could mean 1 in 100 going up to 2 in 100. Was that it? Or was it 25 in 100 jumping up to 50 in 100? This is why we expect stories to quantify harms (and benefits) in absolute terms, not just relative.

It concluded with:

“The American Heart Association recommends men limit themselves to two drinks a day and women to one drink a day, because too much alcohol can raise blood pressure and have other negative effects.”

But again, no information was given on the extent of blood pressure increase or on what “other negative effects” they were talking about. If it’s worth mentioning, it’s worth specifying.

3.  Establish the true novelty of the treatment/test/product/procedure? – NOT SATISFACTORY

We aren’t given any context of the long, long history of research looking at alcohol and heart disease.

4.  Quantify the potential benefits? – NOT SATISFACTORY

The benefit – in terms of risk reduction – was provided only in relative risk reduction terms – 25% reduction in additional heart procedures, heart attacks or strokes. But readers should be told “25% of what?”

5.  Use independent sources and identify conflicts of interest? – NOT SATISFACTORY

No independent expert was quoted – only one of the researchers.

6.  Compare the new approach with existing alternatives? – NOT SATISFACTORY

The potential benefit of alcohol consumption was not compared with any other approach for reducing the risk of additional heart procedures, heart attacks or strokes in people who had bypass surgery. This could have been done with a little homework and a few more sentences.

♥.

* FYI, here’s the full Reuters article with the most misleading bits in blue.

_________________________________________

Sun Nov 14, 2010 4:51pm EST 

“BYPASS PATIENTS CAN BENEFIT FROM A FEW DRINKS

CHICAGO (Reuters) – Men who underwent heart bypass surgery and consumed about two drinks a day afterward had fewer subsequent cardiovascular procedures than those who abstained, according to a study released Sunday.

Light alcohol consumption was associated with a 25 percent reduction in additional heart procedures, heart attacks or strokes in the study by Italian researchers, presented at the annual meeting of the American Heart Association in Chicago.

“The benefit of light amounts of alcohol consumption has been documented in healthy individuals, but our analysis showed a benefit from light alcohol intake in post-coronary bypass patients,” said Dr. Umberto Benedetto of the University of Rome La Sapienza.

However, bypass patients with a condition called left ventricular dysfunction who were heavy drinkers, defined as having more than six drinks daily, were twice as likely to die from heart problems, the study found.

The Italian researchers used a questionnaire to compare alcohol consumption in 1,021 men who underwent heart bypass surgery and reviewed their medical history for 3-1/2 years.

The study also found no adverse correlation between moderate alcohol consumption and any medication.

The American Heart Association recommends men limit themselves to two drinks a day and women to one drink a day, because too much alcohol can raise blood pressure and have other negative effects.

_________________________________________

Read the full Health News Review rating of this Reuters media coverage, and specifically their clear guidelines on how to understand medical research studies.

See also:

2 Responses to ““Alcohol Helps Heart Bypass Patients!” – good news or bad reporting?”

  1. RWT November 29, 2011 at 9:37 am #

    Thanks for this link – helps us to understand how feeble many of these breaking news stories actually turn out to be.

    Like

    • Carolyn Thomas November 30, 2011 at 8:55 am #

      Yes it does. And it also reminds us as consumers to view all ‘breaking news’ in health with a critical eye, asking the same kinds of questions exemplified on Health News Review.
      Cheers,
      C.

      Like

Your opinion matters. What do you think?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: