Guest Articles

Evidence-based medicine and the rational use of diagnostic investigations

Professor Rakesh K. Kumar

Every senior medical student and young doctor want to be able to keep up with the latest advances in medicine. However, the output of published literature keeps rising, so that we are all in danger of drowning in data. It’s difficult enough to keep up with the latest in clinical practice, let alone in basic research.

To at least some extent, evidence-based medicine can help, because it offers approaches that help to turn the data into knowledge which can actually be applied. Notably, these include systematic reviews and meta-analyses, which yield evidence-based practice guidelines that can inform clinical decision-making. Of course, one must remember that guidelines are only generalisations. Achieving the best outcomes for any given patient requires a combination of:

  • skilled clinical observation
  • appropriate investigations
  • application of knowledge and expertise gained by experience
  • the best scientific evidence from the literature.

In this article, I will focus on the appropriate use of investigations. This is an important issue with respect to the care of individual patients, because unnecessary and inappropriate investigations may have adverse effects, while false-positive results may prompt further needless investigation. It is also important with respect to utilisation of resources, particularly in Australia where costs to the health care system are substantially borne by the taxpayer. Over the past decade, the use of laboratory tests has seen a modest annual increase of approximately 3% to 6% [1]. At the same time, requests for diagnostic imaging investigations have increased at approximately 9% per year, so that these services now account for approximately 15% of all Medicare outlays [2].

When looking at evidence-based medicine in the context of the rational use of investigations, it is easy to get lost in the arithmetic of predictive values, probabilities and likelihood ratios. An alternative simpler approach is to rely on the maxim “Only request a laboratory test if the result will change the management of the patient” [3]. This may be an oversimplification in that among other things, investigations are relevant to establishing a diagnosis, excluding differential diagnoses, assessing prognosis and guiding management. Nevertheless, focusing on investigations that matter is sound advice, which is unfortunately all too often ignored.

The quality of the evidence around overuse of diagnostic investigations is relatively low. In hospital settings, however, it has long been recognised that as many as two-thirds of requests for some common Pathology tests may be avoidable, in that they fail to contribute to diagnosis or management [4]. Senior medical students and junior medical officers need to be especially aware of this, because most hospital Pathology test requests are submitted by junior doctors. Among factors that contribute to the uncritical overuse of investigations by JMOs are inexperience, lack of awareness of the evidence base for using a particular investigation and lack of awareness of the cost of the test. Other significant factors are the desire to anticipate the expectations of one’s supervisor and the fear of missing something important. Perhaps the supervisors of PGY1/2 trainees themselves need to drive cultural change and better model the appropriate use of diagnostic investigations!

Some strategies targeted at the test-requesting behaviour JMOs appear to be effective in at least some settings, for example restricting the range of tests that junior doctors may request in emergency departments [5,6]. More generally, management systems with budgetary controls, as well as online systems with decision support, have been promoted [7]. Importantly, education also has a valuable role to play [8].

With funding support from the Commonwealth Department of Health, my colleagues and I developed an open-access website to educate JMOs about the rational use of diagnostic investigations. As a user, you interact with simulated cases and can request investigations as you attempt to establish a diagnosis, while being presented with a running tally of the costs of the tests sought. At the end of each case, you receive feedback via comparison with what an expert would have done. Try it by self-registering, without cost, at The largest collection of cases is targeted to JMOs, but are also likely to be of interest to senior medical students. In addition, there are cases for trainee GPs, plus a few specifically created for advanced trainees in Respiratory Medicine. However, all cases are accessible to all users.

We have evidence that this educational approach can work: in a trial at a large Sydney hospital, we demonstrated that in the period immediately following active engagement of the cohort of junior doctors with this website, there were significant hospital-wide cost savings and an encouraging reduction in the number of blood samples collected from patients [9]. Unfortunately, in agreement with other studies of educational interventions, these changes in test-requesting behaviour were not sustained over the following months. However, there is additional evidence that routine requests for diagnostic investigations can be reduced if junior doctors are provided with cost data at the time of submitting a request [10]. We think a good case can be made for integrating this information into online systems in hospitals, to provide reinforcement.

Meanwhile, I encourage you to have a look at one of the few collections of guidelines about the use of investigations, available on the Australian Choosing Wisely website at These guidelines are supported by a number of specialist medical colleges, notably including the Royal College of Pathologists of Australasia and the Royal Australian and New Zealand College of Radiologists. Also well worth reading is a thoughtful reflection on the “big picture” of overuse and the Choosing Wisely initiative, published late last year and targeted specifically to medical students and trainee doctors [11].



  1. National Coalition of Public Pathology. Encouraging quality pathology ordering in Australia’s public hospitals – Final Report, 2012 (last accessed January 2017).
  2. Australian National Audit Office. Diagnostic Imaging Reforms, 2014 (last accessed January 2017).
  3. Hawkins RC. The Evidence Based Medicine approach to diagnostic testing: practicalities and limitations. Clin Biochem Rev. 2005; 26:7-18.
  4. Hammett RJ, Harris RD. Halting the growth in diagnostic testing. Med J Aust 2002; 177:124-125.
  5. Stuart PJ, Crooks S, Porton M. An interventional program for diagnostic testing in the emergency department. Med J Aust 2002; 177:131-4.
  6. Chu KH, Wagholikar AS, Greenslade JH, O’Dwyer JA, Brown AF. Sustained reductions in emergency department laboratory test orders: impact of a simple intervention. Postgrad Med J 2013; 89:566-71.
  7. Janssens PMW. Managing the demand for laboratory testing: Options and opportunities. Clin Chim Acta 2010; 411:1596-602
  8. Corson AH, Fan VS, White T, Sullivan SD, Asakura K, Myint M, Dale CR. A multifaceted hospitalist quality improvement intervention: Decreased frequency of common labs. J Hosp Med. 2015; 10:390-5.
  9. Ritchie A, Jureidini E, Kumar RK. Educating young doctors to reduce requests for laboratory investigations: opportunities and challenges. Med Sci Educ 2014; 24:161-3.
  10. Feldman LS, Shihab HM, Thiemann D, Yeh HC, Ardolino M, Mandell S, Brotman DJ. Impact of providing fee data on laboratory test ordering: a controlled clinical trial. JAMA Intern Med 2013; 173:903-8.
  11. Lakhani A, Lass E, Silverstein WK, Born KB, Levinson W, Wong BM. Choosing Wisely for medical education: six things medical students and trainees should question. Acad Med 2016; 91:1374-8.