Normalization of a conversation tool to promote shared decision making about anticoagulation in patients with atrial fibrillation within a practical randomized trial of its effectiveness: a cross-sectional study

Abstract

Background Shared decision making (SDM) implementation remains challenging. The factors that promote or hinder implementation of SDM tools for use during the consultation, including contextual factors such as clinician burnout and organizational support, remain unclear. We explored these factors in the context of a practical multicenter randomized trial evaluating the effectiveness of an SDM conversation tool for patients with atrial fibrillation considering anticoagulation therapy. Methods In this cross-sectional study, we recruited clinicians who were regularly involved in conversations with patients regarding anticoagulation for atrial fibrillation. Clinicians reported their characteristics and burnout symptoms using the two-item Maslach Burnout Inventory. Clinicians were trained in using the SDM tool, and they recorded their perceptions of the tool’s normalization potential using the Normalization MeAsure Development (NoMAD) survey instrument and verbally reflected on their answers to these survey questions. When possible, the training sessions and clinicians’ verbal responses to the conversation tool were recorded. Results Our study comprised 183 clinicians recruited into the trial (168 with survey responses and 112 with recordings). Overall, clinicians gave high scores to the normalization potential of the intervention; they endorsed all domains of normalization to the same extent, regardless of site, clinician characteristics, or burnout ratings. In interviews, clinicians paid significant attention to making sense of the tool. Tool buy-in seemed to depend heavily on their ability to see the tool as accurate and “evidence-based” and their perceptions of having time in the consultation to use it. Conclusions While time in the consultation remains a barrier, we did not find a significant association between burnout symptoms and normalization of an SDM conversation tool. Possible areas for improving the normalization of SDM conversation tools in clinical practice include enabling collaboration among clinicians to implement the tool and reporting how clinicians elsewhere use the tool. Direct measures of normalization (i.e., observing how often clinicians access the tool in practice outside of the clinical trial) may further elucidate the role that contextual factors, such as clinician burnout, play in the implementation of SDM. Trial registration ClinicalTrials.gov, NCT02905032. Registered on 9 September 2016.

Explainability for artificial intelligence in healthcare: a multidisciplinary perspective

“…Concerning autonomy , explainability has implications for patients and physicians alike [ 31 ]. One of the major safeguards of patients’ autonomy is represented by informed consent, that is an autonomous, generally written authorization with which the patient grants a doctor his or her permission to perform a given medical act [ 39 ].…”

“…Unlike patient decision aids (which are used by the patient in preparation prior to the clinical encounter), conversation aids are designed for use within the clinical encounter to guide the patient and clinician through the shared decision-making process [28,30]. They incorporate established medical facts about their conditions and, by synthesizing available information, they can help patients to understand their individual risks and outcomes, to explore the available options, and to determine which course of action best fits their goals and priorities [30][31][32]. So, what if individual risk was not calculated using established risk prediction models but instead relied on a validated, yet not explainable, datadriven approach?…”

Supporting translation of research evidence into practice—the use of Normalisation Process Theory to assess and inform implementation within randomised controlled trials: a systematic review

“…Sharpe et al [ 106 ] Canada Qualitative Primary care Paediatric asthma Retrospective 35. Spencer-Bonilla [ 107 ] USA Mixed-methods Cardiology in Hospital Shared decision-making for anticoagulation in atrial fibrillation Prospective 36. Taft et al [ 108 ] Australia Qualitative process evaluation Primary care Long-acting reversible contraceptives Retrospective 37.…”

“…The operationalisation of NPT within the included studies is best categorised as either prospective, in which the initial utilisation of NPT occurs prior to or during the RCT, or retrospective, in which NPT is applied to data that had been previously collected during the RCT after its completion (Table 2 ). The majority of studies, 81% ( n = 34) [ 36 , 68 – 77 , 80 – 91 , 94 , 95 , 97 – 102 , 104 , 105 , 107 , 109 – 114 ], applied NPT prospectively. In exploring the relationship between retrospective or prospective use of NPT and the methodology used in the included studies, out of the 11 studies which used mixed-methods, six [ 69 , 88 – 91 , 98 , 101 , 113 ] applied NPT only to the qualitative aspects of the study and five [ 73 , 81 – 84 , 93 , 102 , 107 ] applied NPT both qualitatively and quantitatively, of which three studies [ 73 , 93 , 107 ] utilised the NoMAD tool.…”

Clinical Ethical Consultation in Decision Making Process

“…However, concern for the patient's condition always helps to reduce his fear. In recent decades, the so-called shared decision-making approach has been widely applied, as shown by the analysis of the present studies [13][14][15][16][17][18][19][20][21][22] . This approach implies not only concern, but also communication and a kind of training of the patient through what he must go through.…”