Article Text

Download PDFPDF
26 Why we should, and how we could, forget about OverDiagnosis and OverTreatment in the clinical encounter
  1. Vije Kumar Rajput1,2,
  2. Mette Kjer Kaltoft3,
  3. Jack Dowie2,3
  1. 1Stonydelph Health Centre, Tamworth, UK
  2. 2London School of Hygiene and Tropical Medicine, London, UK
  3. 3Univeristy of Southern Denmark, Odense, Denmark

Abstract

Research into OverDiagnosis and OverTreatment (OD/OT) may help establish, classify and potentially modify the sources of these phenomena at a population level. However, they represent a distraction from the central clinical task of deciding on the optimal care for the individual person. That task requires combining their personalised preferences with individualised evidence in an evaluation of each adoptable option. We argue that any of three possible ways of introducing evidence on group level OD/OT rates into a clinical decision process - as discrete pieces of information on a single criterion - are unlikely to be helpful. (They will also fail to address the issue of UnderDiagnosis and UnderTreatment (UD/UT) simultaneously.) Firstly, informing the person that there is a ‘possibility’ of OD/OT, perhaps with some qualitative verbal quantification (‘small chance’, ‘moderate risk’) fails to meet the minimum requirements for informed consent, let alone preference-based informed consent. Secondly, including ‘being OD/OT’ (or something similar) as a separate criterion in a multi-criteria Decision Aid is clearly not meaningful, since it will be the consequences of being OD/OT, not being OD/OT, which is of concern. Finally, and the most attempted, introducing numerical rate/s of OD/OT as inputs into a verbal deliberative reasoning process, even when accompanied by well-designed pictograms, are (as we illustrate) likely to result in confusion, misinterpretation or simple abandonment of the attempt to absorb the information and incorporate it coherently into the decision. Fortunately, a Decision Support Tool (DST) based on Multi-Criteria Decision Analysis (MCDA), but lacking any reference to OD/OT, confirms this is a non-issue. In such DSTs, the best available estimates of the performance rates of the available options on all-cause and/or condition-specific outcomes implicitly incorporate the group/population rates of OD/OT – and indeed of UD/UT as well. The problem at the individual level is therefore not of possible over- or under-treatment, but of management out of line with the optimal decision for the person, either because the relevant performance rates are not available, or they are not drawn on, if available. In support of the argument, we introduce the generic MCDA/DST template within which our condition-specific DSTs, such as the one in Bone Health, subsequently presented, can be built. The MCDA structure reduces the effect of knowledge and power asymmetries, as well as being a straightforward way to meet the increasing legal requirement for the informed and preference-based consent of the ‘reasonable patient’. The generic template ensures that individuals (professionals and patients-as-persons) are made aware of the common structure of decisions across all conditions and contexts, having their global decisional competence enhanced through engagement with a specific local decision support tool. Implementation on a widely available spreadsheet platform (Google Sheets) can ensure practical delivery, with minimal resourcing compared with bespoke, condition-specific, decision aids. Many of the professional obstacles and barriers to shared decision making are potentially reduced or avoided.- along with the possibility of OD/OT, or UD/UT.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.