Perspectives
General medicine Information science
Volume 197 - Issue 6

What the public learns about screening and diagnostic tests through the media

Authors:  Amanda J Wilson, Jane Robertson, Benjamin D Ewald and David Henry

Med J Aust 2012; 197 (6): 324-326. || doi: 10.5694/mja11.11504
Published online: 17 September 2012
Better journalistic standards for reporting on medical tests will mean better informed health care consumersScreening of healthy groups in the population and diagnostic testing on suspicion of disease are fundamental components of health care delivery and disease prevention. As the general media are important sources of health information for consumers, accurate and balanced reporting is essential. Media reporting of diagnostic tests is skewed towards screening ...
Download PDF (231 KB)
Share: Share on Blue Sky Share on Facebook Share on LinkedIn Share by email

Better journalistic standards for reporting on medical tests will mean better informed health care consumers

Screening of healthy groups in the population and diagnostic testing on suspicion of disease are fundamental components of health care delivery and disease prevention. As the general media are important sources of health information for consumers, accurate and balanced reporting is essential. Media reporting of diagnostic tests is skewed towards screening tests, particularly cancer screening, and the quality of coverage appears poor.1-3 However, there are no data on how well the media cover stories about diagnostic tests used to confirm disease. The statistics used to quantify diagnostic test accuracy (sensitivity, specificity and the predictive value of positive and negative tests) are difficult to understand, so conveying this information to the public is a challenging task for journalists.3,4 In this article, we review stories written about diagnostic and screening tests in the Australian media, and propose some questions that journalists covering these topics could use.

The quality of media reporting on screening and diagnostic tests

The Media Doctor website rates health news stories using validated 10-item rating instruments.5 The items are consistent with advice provided by the Australian Press Council.6

We identified 1581 medical news stories between June 2004 and February 2011, of which 113 reported on screening tests and 72 on diagnostic tests. Most stories addressed the novelty of the test (80% rated as satisfactory) and avoided disease mongering (90%). Fifty-seven per cent of stories covered diagnostic options, but only 36% discussed the evidence behind the test’s claims, and 24% quantified diagnostic accuracy. Potential harms of testing were covered in only 29%, and 26% mentioned costs. A minority of stories (38%) included independent expert comment to interpret the claims and provide context for the reader. Overall, news coverage of screening tests was superior to that of diagnostic tests (difference in mean proportion of items scored satisfactory, 9%; 95% CI for difference, 3%–15%).

What needs to change?

Our experience, and that of others,1-3,7 suggests seven main areas in which media reporting of diagnostic tests should improve to provide better quality information to the public.

Early phase research

Some of the worst stories of the ones we rated were about early phase or poorly substantiated science. An example is the promotion of thermography for detection of breast cancer in young women (the first story in Box 1). In contrast, the third story highlights the value of good reporting. Pharmaceutical products have a highly regulated and well understood development pathway, but this is not the case with new diagnostic tests. It can be difficult for a non-expert to judge the relevance of the research and the validity of the claims.

Test accuracy

Most stories did not provide quantitative information on the accuracy of the tests, and where this was provided, it was usually confined to their sensitivity (the proportion of cases of the disease or condition that the test detects). Improved sensitivity is often a central claim of stories about new diagnostic tests. However, journalists need to consider the consequences of a wrong diagnosis of disease (false positive), or failing to be diagnosed with a serious disease (false negative). Researchers should talk to the media in terms of the predictive values of a positive or negative test, values which vary with the prevalence of the disease in the population being tested. For example, a positive predictive value (PPV) of 98% means that only 2% of positive results are false positives. One trial of prostate-specific antigen for prostate cancer screening reported a PPV of 32%, meaning that 68% of positive tests in that population were incorrect,12 surely a matter of great public interest.

Harms of testing

Harms are frequently overlooked and include complications of the testing procedure and consequences of false-positive and false-negative results. False positives can lead to anxiety, additional testing, unnecessary procedures and treatments, all with attendant adverse effects.1,2 Screening of healthy populations can create other problems; sensitive tests can detect very early manifestations of disease that may regress, remain subclinical or progress so slowly that it would not be of clinical importance.13,14 Early identification does not always improve clinical outcomes, and patients have died as a consequence of treatment.14

Impact on disease management

Few media reports in our analysis described how a new diagnostic test would influence management of a disease. The second story in Box 1 concerns the identification of individuals at risk of Alzheimer disease. If treatments are not available, people might be told that they have a serious disease with no prospect of treating it, compromising their quality of life.1-3,7 In the case of early detection of Parkinson disease, the fourth story in Box 1 did a much better job. We encourage journalists to ask about the availability of effective treatments for patients who test positive, and whether earlier diagnosis improves clinical outcomes.

Expert advice

Because the methodological and statistical concepts in diagnostic testing can be challenging,3,4 it will often be helpful if journalists ask an independent expert to comment on the quality of the research behind claims about new tests and to interpret research findings. We think clinical epidemiologists should make themselves available for this task, perhaps by being rostered to science media centres that are developing in some countries (eg, http://www.sciencemediacentre.org and http://expertguide.com.au). We recognise that modern media staff work to short deadlines and content experts will have to give high priority to these requests.

Genetic testing

Journalists need to be particularly cautious when writing stories about genetic testing, and should always seek independent expert opinion. Genetic testing is an increasingly important field with tests being promoted directly to the public. In the United States, tests for around 480 traits are now on offer.15 Many of these are claimed to be predictive of characteristics other than susceptibility to disease (eg, creative, linguistic, musical or athletic abilities). Marketers of genetic tests appear to frequently misrepresent their utility, as the science of genetic testing is complex and initial research findings are often not reproducible.15

Costs

Our review found that the majority of the stories did not adequately address the costs of diagnostic testing. These may fall on the health care system where it is publicly funded, or on individuals or their insurance providers. Some genetic tests are promoted directly to consumers and may cost hundreds of dollars. It is important to understand the budgetary impact of a new test or screening program and where this will be felt.

Question journalists should ask

Box 2 lists questions that we recommend journalists ask when presented with claims about new tests. If the research is in the preclinical phase it is important that this is made clear to journalists and readers (question 1). If the test is currently used or under clinical testing, the journalist should ask about its value compared with any existing tests, the level of evidence supporting it, the type of test (screening or diagnostic) and the rate of false-positive and false-negative results (questions 2–7). Journalists should consider the adverse consequences of applying diagnostic tests to low-risk populations (questions 5–7). From our experience, stories about diagnostic tests are often based on new research findings. Informants may adopt an advocacy position, emphasising the superior accuracy of the test or other advantages, such as simplicity and convenience, and downplaying other aspects. Questions 2–5 deal with claims of improved diagnostic accuracy and real benefits to patients. It is difficult for most people to evaluate the methodological quality of studies of diagnostic test accuracy. Reports of good studies will comply with the Standards for the Reporting of Diagnostic Accuracy guidelines, and journalists should ask their informant whether this is the case (question 6).16

Conclusions

Consumers are increasingly involved in clinical decisions and need access to accurate and balanced information on their options. This includes screening and diagnostic choices as well as treatments. The media play a crucial role here, and in our view are not fulfilling it adequately. Inaccurate information about diagnostic, screening and genetic tests can mislead patients into undergoing unnecessary testing with consequent costs and adverse effects. This can only happen with acquiescence of doctors, who should be aware of the limitations of media coverage of diagnostic tests.

1 Examples of good and bad reporting of screening and diagnostic tests in the Australian media

Headline

Media outlet

Claim

Criticisms

Score*

New scanner aims to detect breast cancer in young women

The West Australian (Seven West Media)8

“In an Australian first, WA researchers are testing a simple scanner that could allow GPs to pick up early breast cancers, particularly in younger women who cannot have mammograms”

Inappropriate promotion of untested technology (thermography) for a low-risk group without considering consequences of false-positive or false-negative results.

10%

Alzheimer’s can be predicted with “100 per cent accuracy”

The Australian (News Corporation)9

“Alzheimer’s disease can be predicted with up to 100 per cent accuracy years before its onset using biomarkers found in spinal fluids, a study says”

Reported sensitivity only; no reference to false-positive rates, consequences of false diagnoses or lack of effective treatment.

22%

Breast “tests” could offer false sense of security

Sydney Morning Herald (Fairfax Media)10

“Cancer specialists warn that private clinics offering unproven breast screening methods as a ‘safe’ alternative to mammograms could be putting women’s lives at risk”

Pointed out lack of evidence for thermography in comparison to mammography. Strong independent comment highlighted consequences of missed diagnosis.

89%

Parkinson’s test warning

Herald Sun (News Corporation)11

“Melbourne researchers have developed a breakthrough blood test that can detect Parkinson’s disease, even in those with no symptoms”

Quantified test accuracy, spelled out possible adverse effects, addressed availability and importance of effective treatment in early diagnosis.

100%

* Scored according to Australian Media Doctor rating instruments. WA = Western Australia. GP = general practitioner.

2 Questions journalists should ask about new tests

STARD = Standards for the Reporting of Diagnostic Accuracy.

Authors

  • Amanda J Wilson PhD1
  • Jane Robertson PhD1
  • Benjamin D Ewald BMed, MMS, PhD1
  • David Henry FRCP2
  • 1. University of Newcastle, Newcastle, NSW.
  • 2. Institute of Clinical Evaluative Sciences, Toronto, Ontario, Canada.

Correspondence: amanda.wilson@newcastle.edu.au

Competing interests

No relevant disclosures.

Acknowledgements

We thank Marc Bevan for major contributions to this article through data collection and analysis, writing and editing.

References

Provenance: Not commissioned; externally peer reviewed.

View this article on Wiley Online Library

More like this

General medicine Perspective 17 November 2025 Open Access

The CURE Asthma roadmap

Gary P Anderson, Anthony Flynn, Phil G Bardin, John D Blakey, Shyamali C Dharmage, Paul Foster, Peter G Gibson, Adam Jaffe, Alan James, Christine R Jenkins, Sundram Sivamalai, Peter D Sly, Guy B Marks, Vanessa M McDonald, Judy Wetttenhall

View more