Science News

General Health Checks Generally Unhelpful

Sun, 25th Nov 2012

Ben Valsler

Listen Now    Download as mp3 from the show Investigating ISIS - The Neutron Source

A systematic review of research into general health checks shows that, although they increase the number of diagnoses, they do not reduce the number of deaths from cardiovascular disease or cancer.

Routine health checks, or testing healthy people for indicators of a range of conditions, are common elements of healthcare in some countries.  The idea is that, by checking regularly for risk factors such as high blood pressure or high cholesterol, we should be able to reduce morbidity and prolong lives.  It seems a very worthwhile aim, and intuitively we feel that health checks should be a good thing.

However, as we discussed with regards breast cancer screening a few weeks ago, there is always a risk of over-diagnosis – finding evidence of a condition that would not have become a problem, and potentially therefore recommending unnecessary treatments.

Little research has been done to see if health checks actually achieved their aims.  So now, a team at the Nordic Cochrane Centre in Copenhagen led by Dr Lasse Krogsbøll combined data from 14 studies into a meta-analysis to examine the pros and cons of health checks.

Publishing in the British Medical Journal, Krogsbøll and colleagues looked at mortality, morbidity, number of diagnoses, hospital admissions, referrals to specialists and a number of other factors indicating health, including self-reported health, worry and absence from work.  They used this to compare screened populations with those that received no intervention, in studies with between four and twenty two years follow up.

Although the results varied between studies, this meta analysis found no evidence in the health-checked group for a reduction of either total mortality, or deaths through cardiovascular disease or cancer.  There was evidence for an increase in diagnosis, but as this didn’t lead to a change in mortality, we can assume that these are often over-diagnoses.

So what does this say for general health checks?  The authors argue that these health checks may appeal to people who need them less, and point out that doctors already carry out screening in patients where they think it’s needed, and conclude that health checks “do not work as intended”.  In a related Editorial piece, Domhnall MacAuley suggests that “Policy should be based on evidence of wellbeing, rather than on well meant good intentions.”



Subscribe Free

Related Content


Make a comment

Most interesting.
I've always wondered if a general physical exam was necessary for individuals who are otherwise generally healthy when a large part of the study is to ask if the patient has any "symptoms". 

If one runs a panel of lab tests, it is not uncommon to find a couple of values slightly outside of the "normal" range.  Then what should one do with these results?  It tends to cause unnecessary worry for the patients. 

There has been a lot of debate about prostate screening.  Metastatic prostate cancer can be very devastating.  However, a large number of 80 yr old men have asymptomatic prostate cancer contained entirely within the organ.

I would wonder, however, if this study should be separated by individual test.  For example, there may be benefits of fecal occult blood testing, and periodic colonoscopy, and early removal of polyps which might otherwise be undetected.  CliffordK, Sun, 25th Nov 2012

Having a camera stuck where the sun don't shine is surprisingly expensive ...

So there would be enormous cost for everyone to have one done annually.

There is also a small risk of serious complications of colonoscopy, (e.g. bowel perforation).
RD, Mon, 26th Nov 2012

Not annually... 
But, perhaps at 50, 60, 70, or so.

Sigmoidoscopy is less invasive than colonoscopy, but has the potential to miss cancer and precancer.

Good point about the risks.  According to the article, somewhere about 0.2% to 0.6% of the colonoscopies result in perforation, with a resulting 3% to 10% mortality.

Of course not all polyps develop into cancer, but it is far cheaper, with better outcomes when treated early.

Overall each test should be evaluated for its efficacy.  Even listening to the heart once a year adds up.  What is better, $100/yr to listen to the heart for a decade, or 1 colonoscopy per decade? 

A physician's assistant, or even a nurse could do routine blood pressure screening.  Or, even self administered tests.  Blood pressure, of course, is something that needs some monitoring in some individuals.

The HPV vaccine could reduce things like the need for a PAP smear. CliffordK, Mon, 26th Nov 2012

This was certainly what they recommended for future research - looking at individual tests, rather than blanket screening. BRValsler, Tue, 27th Nov 2012

See the whole discussion | Make a comment

Not working please enable javascript
Powered by UKfast
Genetics Society