Decision making about healthcare–related tests and diagnostic test strategies: what “The death of Ivan Ilyich” teaches us

Abstract

In the last few years, I have been interested in the field of diagnostic imaging evaluation and, like others in this field, I have been influenced by the 6-step model proposed by Fryback and Thornbury [[1]Fryback D.G. Thornbury J.R. The efficacy of diagnostic imaging.Med Decis Making. 1991; 11: 88-94Crossref PubMed Scopus (965) Google Scholar]. This model has been a cornerstone in this area, and many papers have been published discussing how to apply it. As an example of the influence of this model, last year, the Journal of Clinical Epidemiology published a series of papers on diagnostic test evaluations [2Schünemann H.J. Mustafa R.A. Decision making about healthcare-related tests and diagnostic test strategies. Paper 1: a new series on testing to improve people’s health.J Clin Epidemiol. 2017; 92: 16-17Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar, 3Mustafa R.A. Wiercioch W. Cheung A. Prediger B. Brozek J. Bossuyt P. et al.Decision making about healthcare-related tests and diagnostic test strategies. Paper 2: a review of methodological and practical challenges.J Clin Epidemiol. 2017; 92: 18-28Abstract Full Text Full Text PDF PubMed Scopus (23) Google Scholar, 4Mustafa R.A. Wiercioch W. Falavigna M. Zhang Y. Ivanova L. Arevalo-Rodriguez I. et al.Decision making about healthcare-related tests and diagnostic test strategies. Paper 3: a systematic review shows limitations in most tools designed to assess quality and develop recommendations.J Clin Epidemiol. 2017; 92: 29-37Abstract Full Text Full Text PDF Scopus (6) Google Scholar, 5Mustafa R.A. Wiercioch W. Arevalo-Rodriguez I. Cheung A. Prediger B. Ivanova L. et al.Decision making about healthcare-related tests and diagnostic test strategies. Paper 4: International guidelines show variability in their approaches.J Clin Epidemiol. 2017; 92: 38-46Abstract Full Text Full Text PDF PubMed Scopus (8) Google Scholar, 6Mustafa R.A. Wiercioch W. Ventresca M. Brozek J. Schünemann H.J. DU-Diagnosis expert group. Decision making about healthcare-related tests and diagnostic test strategies. Paper 5: a qualitative study with experts suggests that test accuracy data alone is rarely sufficient for decision making.J Clin Epidemiol. 2017; 92: 47-57Abstract Full Text Full Text PDF PubMed Scopus (8) Google Scholar] and the 6-step model was one of the bases for the analyses performed. Also, in a more recent article published this year in the Journal of Clinical Epidemiology, this model was again one of the bases of the analyses [[7]Whiting P. Leeflang M. de Salis I. Mustafa R.A. Santesso N. Gopalakrishna G. et al.Guidance was developed on how to write a plain language summary for diagnostic test accuracy reviews.J Clin Epidemiol. 2018; 103: 112-119Abstract Full Text Full Text PDF Scopus (7) Google Scholar]. However, although the Fryback and Thornbury model is fundamental for the comprehension of the whole process of diagnostic test evaluation, it has, in my opinion, a philosophical bias in its analytical structure. This bias occurs from step 4 ahead when it considers a diagnostic test useful only if it changes the therapeutic process of the patient and if this change leads to an increment in the amount or quality of life at an acceptable cost. While these assumptions are valid in some situations, as in the evaluation of screening programs, or when the analyses are performed under the perspective of the payers, from the patients' and physicians' perspectives in a diagnostic or prognostic evaluation, to use a diagnostic test only when it modifies the therapeutic strategies and the outcome of the patients can be an object of deserved criticism. A classical example is the genetic testing for Huntington's chorea. Although the use of this test for general population screening is hardly justifiable due to the rarity of the disease and the absence of effective therapies, its use in patients with clinical symptoms suggestive of the disease could be important to confirm or exclude the diagnosis [8Craufurd D. MacLeod R. Frontali M. Quarrell O. Bijlsma E.K. Davis M. et al.Diagnostic genetic testing for Huntington’s disease.Pract Neurol. 2014; 15: 80-84Google Scholar, 9Margolis R.L. Diagnosis of huntington disease.Clin Chem. 2003; 49: 1726-1732Crossref PubMed Scopus (82) Google Scholar]. Here, it is important to say the obvious: to know that there is nothing to do for a medical condition first, it is necessary to know what is causing that medical condition. Therefore, in clinical practice, it is often the case that the diagnosis of a medical condition is important by itself and an illustration of this importance can be obtained from the literary fiction. In the Leo Tolstoy's The Death of Ivan Ilyich [[10]Tolstoy L.N. The Death of Ivan Ilych, translation Louise and Aylmer Maude. Christian Classics Ethereal Library.http://www.ccel.org/ccel/tolstoy/ivan.pdfDate: 1886Date accessed: August 20, 2018Google Scholar], the drama of a patient who does not know the cause and the prognosis of his medical condition is brilliantly described and can be synthesized in the following paragraphs of the book:He said nothing of this, but rose, placed the doctor's fee on the table, and remarked with a sigh: “We sick people probably often put inappropriate questions. But tell me, in general, is this complaint dangerous, or not?…”“I have already told you what I consider necessary and proper. The analysis may show something more.” And the doctor bowed.Ivan Ilyich went out slowly, seated himself disconsolately in his sledge, and drove home. All the way home he was going over what the doctor had said, trying to translate those complicated, obscure, scientific phrases into plain language and find in them an answer to the question: “Is my condition bad? Is it very bad? Or is there as yet nothing much wrong?” And it seemed to him that the meaning of what the doctor had said was that it was very bad. Everything in the streets seemed depressing. The cabmen, the houses, the passers-by, and the shops, were dismal. His ache, this dull gnawing ache that never ceased for a moment, seemed to have acquired a new and more serious significance from the doctor's dubious remarks. Ivan Ilyich now watched it with a new and oppressive feeling. In conclusion, although the Fryback and Thornbury 6-step model remains an important and very useful guide in the field of diagnostic test evaluation, we cannot restrict the value of a diagnostic test to situations where it changes the therapeutic process and the outcome of the patients. It is also important that the impact of the diagnosis per se on the patients' well-being would be considered and the analytical models should evolve to contemplate this essential aspect of the diagnostic process. Download .xml (.0 MB) Help with xml files Data Profile