Issue 19 / 30 May 2011

UNPROVEN screening tests increase medication use, health care costs and anxiety but produce little clinical benefit, according to experts commenting on research in the Archives of Internal Medicine.

Researchers compared the clinical course in 1000 asymptomatic patients who underwent coronary computed tomographic angiography (CCTA) screening, with that in 1000 similar patients who did not have the screening test. (1)

Over an 18-month follow up, patients with an abnormal CCTA result were more likely to use statins and aspirin and to undergo coronary revascularisation procedures than the controls. They were also more likely to be referred for secondary tests, despite their asymptomatic status and low Framingham Risk Score.

However, the rates of cardiovascular events were identical — and very low — in both groups.

“We found that evidence-free performance of CCTA in asymptomatic patients was associated with further evidence-free testing and interventions”, the researchers said.

Cardiac surgeon Professor James Tatoulis, who is chief medical advisor at the National Heart Foundation, said that his organisation does not recommend CCTA screening.

“Although people with high calcium scores on CCTA have a higher possibility of having cardiovascular disease, knowing this does not seem to alter their medium-term outcomes”, he said.

“It may seem intuitive to screen patients, and act on positive findings … but this is not necessarily borne out by the evidence”, he said.

An editorial in the Archives of Internal Medicine said the research was a “powerful reminder of the 2-edged effects of screening”. (2)

The editorial, titled “Pseudodisease, the next great epidemic in coronary atherosclerosis?”, emphasised that overdiagnosis led to increased tests, medications, or even surgical procedures, all of which carried risks.

“Over diagnosis is threatening to become an increasingly important public health problem because of the enthusiasm for and proliferation of unproven screening tests”, the editorial said.

Professor Janet Hiller, professor of public health at the Australian Catholic University, said there was a range of unproven screening tests available to the Australian public, sometimes marketed over the internet.

“I think we need to look at whether our regulatory approaches have kept up with the marketing”, she said.

Professor Hiller said non-evidence-based screening may cause harm beyond the immediate patient, because people who received a positive result often presented at their GP seeking follow-up tests.

“So the companies make the profit but it’s our public health system that has to wear the cost”, she said.

Professor Stephen Leeder, director of the Menzies Centre for Health Policy at the University of Sydney, said screening had been controversial for decades.

“Screening is notorious for turning people into patients,” he said.

“Unless you have unequivocal evidence of screening detecting a real precursor of clinical disease at a point where treatment is effective, and you can assure compliance with that treatment … you generate useless pain and worry and huge expense,” he said.

The Archives study authors and editorial called for randomised trials to further examine the benefits of heart disease screening.

“… we know that many people die with, rather than because of, [heart disease]. If we are going to prevent an epidemic of coronary pseudodisease, we as a profession will have to muster the courage, imagination, and discipline to design and perform the needed large-scale trials,” the editorial said.

– Sophie McNamara

1. Arch Intern Med 2011 (Published online May 23)

2. Arch Intern Med 2011 (Published online May 23)

Posted 30 May 2011

9 thoughts on “Pseudodisease: a risk of screening?

  1. JD says:

    @Sue Ieraci:
    The paper is not about calcium scores it is about CT angiography of coronary arteries:
    “Calcium scores were measured as part of the prior study protocol,but because these scores were not reported to the physician or patients, they have not been included in our analysis of downstream behavior.”
    The paper and the MJA Insight might be appropriately critical about screening tests, but the data used is not good, and the MJA Insight article should come up with a better example and correctly represent what the article is talking about, which is not calcium scores (as Prof Tatoulis’ quoted comment might suggest).
    CT angiography is non-invasive and more informative but we need more data to know what its place is in screening. A proper RCT outcome study in a population representative of patients we see is needed (ie, not well Koreans). At the moment, cardiologists get paid for doing stress tests & echos (with or without stress) as screening tests for the stream of people going through their offices with chest pain, a large proportion of which is non-cardiac and probably diagnosable by taking a decent history; this is at significant expense to the public purse. Bayes theorem will also tell you that occasionally this approach will miss patients who have significant IHD.

  2. Sue Ieraci says:

    This paper is not about all types of screening. There are two significant issues here. One is that this was a test applied to a low risk (asymptomatic) population. According to Bayes theorem, this is a poor use of a test, and likely to maximise the false positives. Secondly, it is known that coronary calcification (detected on CT) is very common and does not correlate with coronary occlusion or clinical outcomes. The paper is appropriately critical about the use of a non-specific test on a low-risk population. This does not say anything about all other screening tests.

  3. JD says:

    This a really poor study: Anon points to the studies own comments of its limitations.
    MJA Insight is right to point out the limitations of screening tests, but come up with a better example.
    @bewildered middle aged doc: you should know that there are NO long-term clinical outcome studies on obesity management. There are some short-term studies using surrogate markers but this is far from being solid evidence. The data for lipid lowering in high-risk patients is solid.

  4. dr roger burgess says:

    Dream on “bewildered middle age doc”. Thank goodness people are quite rightly concerned about their cholesterol levels. You should embrace the massive evidence of the benefits of lowering it to levels “recently re-set even lower”. I must agree with the comments of earlier commentators about the pathetic short-termism of this article, emanating from such a prestigious journal as the Archives of Internal Medicine. This sort of study should be of Framinghamesque proportions to have any relevance to modern medical reality. Patients are increasingly seeking answers, particularly when they they see friends and relatives drop dead in front of their eyes. ONE THIRD of all first infarcts are FATAL. We can’t scan these poor devils!

  5. bewildered middle aged doc says:

    I have been interested to notice while taking clinical histories, the number of patients’ relatives who suffer from this peculiar disease “cholesterol” and are being industriously treated (while completely asymptomatic) with powerful drugs that have potentially serious side effects. The aim seems to be to reduce their ‘levels’ of this dreadful substance to under some arbitrary point (recently re-set even lower so that a large proportion of the population are ‘at risk’). Perhaps some of the massive expenditure on these drugs could be redirected to smoking, reduction, obesity management, treatment of hypertension and promotion of exercise and healthy eating (all of which have evidence to back them up).

  6. Anon says:

    The study itself points out many limitations – it was a “non-randomized study of self-referred patients and is thus subject to allocation bias, selection bias, and residual confounding. Specifically, the control group did not opt for CCTA screening and may have also been less likely to fill prescriptions and undergo any recommended secondary testing. We attempted to minimise these sources of bias by matching and adjusting for other variables in the multivariate logistic regression models. This was a racially homogenous Korean population, which may limit external generalization. Our study is also limited by a small number of events and relatively short follow-up (18 months), which are insufficient to form conclusions about cardiac events in a low-risk cohort.”
    The conclusions of the researchers were far more nuanced than the article indicates and suggested that further studies were required.
    It would be good if the AMA provided a more scientific summary pointing out the limitations of trials as well as the findings rather than just a journalistic type article more suitable for general mass media

  7. John Stokes says:

    Below are the WHO recommendations for screening. Outlines the important considerations before we embark on these programs which divert money from care to corporations and SODs (Single Organ Doctors).
    WHO recomendations:
    The condition should be an important health problem.
    There should be a treatment for the condition.
    Facilities for diagnosis and treatment should be available.
    There should be a latent stage of the disease.
    There should be a test or examination for the condition.
    The test should be acceptable to the population.
    The natural history of the disease should be adequately understood.
    There should be an agreed policy on whom to treat.
    The total cost of finding a case should be economically balanced in relation to medical expenditure as a whole.
    Case-finding should be a continuous process, not just a “once and for all” project.

    Says it all.

  8. yc says:

    With only 18 months follow up it is hard to come to your conclusion that screening is a complete waste of time.

Leave a Reply

Your email address will not be published. Required fields are marked *