Should you skip your next check-up?

Have you had your annual physical yet this year?  It’s been a while, hasn’t it.  Come now, we’re all friends here: you can be honest.  You’ve put it off for several years now, haven’t you?  Perhaps you feel a little bit guilty about it?  You do know you should go, though.  Right?  It’s very important that you see your medical provider every year for helpful advice and disease detection, no matter who you are or how well you feel.

Except maybe not?  From the New York Times:

Among physicians, researchers and insurers, there is an ongoing debate as to whether regular checkups really reduce the chances of becoming seriously ill or dying of an illness that would have been treatable had it been detected sooner.

No one questions the importance of regular exams for well babies, children and pregnant women, and the protective value of specific exams, like a Pap smear for sexually active women and a colonoscopy for people over 50. But arguments against the annual physical for all adults have been fueled by a growing number of studies that failed to find a medical benefit.

What?!?  Nooooooooooo!  You’ll have to pry annual check-ups from our cold, dead hands.  Is nothing sacred?  The next thing we know, you’ll be coming for our fancy head mirrors.

I should pause in my glibness and note that I’m getting off the hook easy here.  As the quote notes, the benefit of regular check-ups for babies and children is well-established, thus the particulars of my own practice aren’t under the microscope here.  Further, I performed precious few annual screening exams on adult patients even during my training, and haven’t laid hands on a grown-up patient for a decade and a half, so I’m hardly in a position to comment authoritatively.  Providers who actually perform routine physicals on adult patients might well have a different take.

But with that out of the way, I do think it’s important that medicine continue to examine its practices and question if what it provides is of genuine value. No matter how cherished some part of our current standard of care may be, if there isn’t evidence to support its benefit to the patient, then it should be jettisoned.

However, I think the Times article creates an inappropriately conclusive impression.  Perhaps other readers will disagree, but my take-away is that there’s convincing evidence annual physicals don’t provide any real benefit in terms of preventing patient mortality, and thus people can start skipping them without any negative consequence to their longevity.  Looking more closely at the studies the article cites, I’m not so sure we’re really there yet.

Dr. Ateev Mehrotra of the University of Pittsburgh School of Medicine, who directed a study of annual physicals in 2007, reported that an estimated 44.4 million adults in the United States undergo preventive exams each year. He concluded that if every adult were to receive such an exam, the health care system would be saddled with 145 million more visits every year, consuming 41 percent of all the time primary care doctors spend with patients.

There is already a shortage of such doctors and not nearly enough other health professionals — physician assistants and nurse practitioners — to meet future needs. If you think the wait to see your doctor is too long now, you may want to stock up on some epic novels to keep you occupied in the waiting room in the future.

I will commend the Times for linking to full-text articles, rather than simply abstracts.  By doing so, it allows readers to look at the science themselves and come to their own conclusions rather than simply taking the writer’s analysis at face value.  It is a more intellectually honest approach, and I appreciate that.

And in fact I am questioning how the writer is presenting these and other data.  The above linked study isn’t really about how beneficial preventive visits are (though it does mention that this question is a controversial one), but merely how and why they were performed.  Its author does note the stress that performing annual preventive visits for every adult as currently recommended would place on the supply of healthcare providers, but that’s a different question from whether or not the visits are actually worthwhile.  The study does not draw a conclusion with regard to the latter issue.  Indeed, from elsewhere in the study’s conclusions:

While many physicians believe that a preventive visit is an important mechanism to detect subclinical illness and to improve physician-patient relations,5– 6 we do not know how frequently PHEs and PGEs fulfill these roles. [emphasis mine, here and below]

Detecting subclinical illness is a major reason for performing annual physicals, but this study doesn’t speak to that question.  Before we seriously consider sending annual physicals the way of the iron lung, we need to settle it better, right?

Hence the second (and only other) study cited:

The research team, led by Dr. Lasse T. Krogsboll, analyzed the findings of 14 scientifically designed clinical trials of routine checkups that followed participants for up to 22 years. The team found no benefit to the risk of death or serious illness among seemingly healthy people who had general checkups, compared with people who did not. Their findings were published in November in BMJ (formerly The British Medical Journal).

In introducing their analysis, the Danish team noted that routine exams consist of “combinations of screening tests, few of which have been adequately studied in randomized trials.” Among possible harms from health checks, they listed “overdiagnosis, overtreatment, distress or injury from invasive follow-up tests, distress due to false positive test results, false reassurance due to false negative test results, adverse psychosocial effects due to labeling, and difficulties with getting insurance.”

Furthermore, they wrote, “general health checks are likely to be expensive and may result in lost opportunities to improve other areas of health care.”

In summarizing their results, the team said, “We did not find an effect on total or cause-specific mortality from general health checks in adult populations unselected for risk factors or disease. For the causes of death most likely to be influenced by health checks, cardiovascular mortality and cancer mortality, there were no reductions either.”

Let’s start with how this information is presented.  Yes, the article’s author does write that the quote comes “in introducing their analysis,” but I think the subtlety of that would be lost to most readers.  (Feel free to disagree.)  The bulk to the material in the quoted passage is from the introduction to the study, not the conclusion.  They are possible harms, not harms that the study actually found.  They are posited by its authors, not soundly demonstrated.  I think the Times article makes it seem otherwise.

Furthermore, this study is a meta-analysis, which is basically a study that studies a bunch of other studies.  The axiom I learned in my training about such meta-analyses is “garbage in, garbage out.”  In other words, a meta-analysis is only as good as the studies it examines.  And I think there are major flaws that the Times article elides.

In particular, a reader might conclude that her risk of dropping dead of a heart attack is no lower if she gets annual check-ups or doesn’t.  That’s what I’d take away from “no reductions in mortality,” right?  But look at this from elsewhere in the study:

There are several possible explanations for this. Most of the trials were old and consequently used treatments different from what would be used today—such as clofibrate or nicotinic acid for hypercholesterolaemia, instead of statins. Also, thresholds for treating cardiovascular risk factors were higher than they are today. However, it is not a given that the results would be better today, as medical innovations sometimes prove harmful29 and as reducing risk factor thresholds means treating people at lower risk who have a smaller potential for benefit but the same risk of harm.

The bolded bit is the garbage in.  If the majority of the studies analyzed for benefit in reducing cardiovascular mortality used outmoded and abandoned treatments as a risk-reduction mechanism, then it is not sound to use them in drawing conclusions about current medical practice.  Period.  The authors try to hedge by talking about thresholds for treatment and the potential for harm from such treatments, but I find their reasoning speculative at best.

But what about cancer?

Finally, some of the trials used only one health check instead of repeated health checks. For cancer mortality, subgroup analysis showed a trend towards benefit from more than one health check and towards harm from one health check only. For cardiovascular mortality, the opposite trends were observed. We regard these results as chance findings. Also, it is not a given that several health checks would be better than one, as some of the harms would increase.

If I’m reading that correctly, it means that for studies where only one check-up was done, a trend toward harm from the visits (such as overdiagnosis, let’s say) was found.  If there were more than one check-up included in the trials, then there was a trend toward benefit.  The authors dismiss that as chance, but I don’t think it’s correct to do so.  One could quite reasonably conclude that the benefit from cancer detection accrues over time, rather than from a one-time spot check.  To my reading, the conclusion that preventive visits confer no benefit in cancer mortality is unsound.

So if the lack of benefit in reducing cardiovascular mortality can be accounted for by the outmoded interventions included in the trials, and if subgroup analysis can be interpreted differently than the authors of the study choose to report, then perhaps there’s still some reason to get preventive checks after all?

Looking over this post, I fear that it is far too dry and abstract to be enjoyable for anyone to read.  (I apologize to those of you who catch yourselves nodding off while reading it.)  Let me put it in these terms — though I don’t have circulation numbers for the respective publications, I’ll wager the readership of the New York Times vastly surpasses that of the BMJ.  Further, the former comprises primarily lay readers, while presumably only physicians and other healthcare professionals who are used to reading medical literature critically bother with the latter. I therefore feel that the nation’s paper of record, which is viewed as a source of reliable medical information by a great many people, has an obligation to report medical findings judiciously and with as balanced an interpretation as possible.

As it stands, readers of the Times could come across the linked article and decide that there is an emerging consensus that annual physical exams don’t really offer a real preventive benefit, and that they can go ahead and skip it.  I’m not an internist or family practitioner, so I am not abreast of what the preponderance of reliable evidence really is, but if you scrutinize the only two studies the Times author uses as basis for her piece, you see that it’s not nearly so cut and dried as (at least I think) she makes it seem.

If there really is a benefit in getting an annual physical (which I think is still quite plausible), then this piece’s overly broad tone with a very narrow base is a disservice to its readers.  Perhaps there really isn’t as much benefit as was once thought, but that case isn’t made in this essay nearly as much as the author seems to imply.

Russell Saunders

Russell Saunders is the ridiculously flimsy pseudonym of a pediatrician in New England. He has a husband, three sons, daughter, cat and dog, though not in that order. He enjoys reading, running and cooking. He can be contacted at blindeddoc using his Gmail account. Twitter types can follow him @russellsaunder1.


  1. Great post, and an interesting article by the Times.

    I tend to like what my doctor does. There are a few things that, because of age and family history, he does some minor checking up on each year. Mostly, though, he focuses the annual exam on a more holistic wellness consultation than most other doctors I’ve seen. There’s a lot of talk about things like stress, nutrition, exercise, etc. In fact, my annual exam is about 9 parts that and 1 part “Say ahhhh.”

    It might well be that this kind of annual checkup doesn’t produce results for most people, but for me it’s goos to have a regular refocus on my lifestyle choices with someone trying to help me live longer and more able.

  2. I went for a checkup shortly after turning 40. My doctor said to come back when I turn 45, or when there’s a problem. That didn’t sound right so I confirmed, “Don’t you want to have me on annual checkups? I’m not old, but I’m not young anymore either.” “You ain’t broke, so there’s nothing for me to fix,” he said.

    • To be sure. I’m not saying that the evidence is conclusive that annual physicals are necessary. I’m just miffed that an article in the Times makes it seem pretty conclusive that they aren’t, based on misrepresenting the findings of one study and the shaky findings of a second one.

      • True, this may be attitudnal on my doctor’s part — he clearly sees his role as one of healing, providing remedy, so apparently he is dismissive of the notion that healthy-patient checkups provide substantial prophylaxis for other kinds of problems, at least for heavy-but-otherwise-healthy men in their early 40’s.

    • I’m not a doctor, but I’m pretty sure most responsible physicians will tell you that PCP is not only strongly addictive but also deletorious to your physical and mental health.

  3. The most I’ve seen my PCP for (post baby) is vaccines and “women’s wellness” exams. I was surprised to hear that they only wanted to do a pap every other year. They do ask if they can test for certain STDs, which I found odd when I was married. But whatever, I’ve decided not to take offense.

  4. One of the things about meta analysis that’s trickier than “garbage in, garbage out” is “are you actually reviewing the literature covering the entire domain”. If you’re looking at multiple studies, but you pre-select those studies which might support your conclusion and not pay too much attention to the ones that don’t, you have done very bad meta analysis.

    • I’m not familiar enough with the literature on the question at hand to know how well it was covered by the meta-analysis cited. However, every time I go back and re-read it I find another manifest flaw.

Comments are closed.