**1. Introduction**

Decision-making is a complex and progressively unpredictable process that relies on precise information availability (Simon 1947; Sousa et al. 2019).

In healthcare, the decision-making process called the "clinical decision process" (Higgs et al. 2019) is a part of a more complex process named "clinical reasoning". According to Norman (2005), the topic of clinical reasoning has been studied for more than 30 years and three fields of research have characterised it. They are related to (i) Understanding the process of clinical reasoning; (ii) Knowledge and memory related to clinical reasoning; and (iii) Mental representation of clinical reasoning.

Up to now, there is no unique definition of clinical reasoning, there are several definitions that differ a lot from each other (Durning et al. 2013). For clearness, in healthcare literature often clinical reasoning and clinical decision-making have been used as synonyms to define "the process by which a healthcare practitioner decide what to think and do with a patient" (Christensen et al. 2017, p. 176), but Durning et al. (2013, p. 446) provides a precise definition of clinical reasoning as "the mental process and behaviours that are shared (or evolve) between the patient, physician, and the environment". In this manuscript, according to Higgs et al. (2019), we refer to clinical reasoning as the whole process of a physician's thinking during her/his practice, where clinical decision-makingis a step ofit, providedin order to emphasize the outputs or decisions. Towards the analysis of every decision-making process, it should

be noted that this is a human process and it is not exempt from human characteristics such as several biases, that is, deviations from human rationality, due to human nature (Kahneman 2011). Most of the literature on clinical reasoning (both articles and reviews) leads to bias-triggered diagnostic errors (Cooper and Frain 2016) concerning emergency medicine (Croskerry 2003; Rubio-Navarro et al. 2020; Antonacci et al. 2020); just a few studies, however, regard biases in the clinical and therapeutic decision process (day by day decisions). This first literature gap would deserve to be further investigated; especially, some scholars have stated that clinical outcomes may be improved through recognising, understanding, and modifying those decisions a ffected by biases. (Antonacci et al. 2020).

Moreover, some authors outside the healthcare field (Baron 1998; Haley and Stumpf 1989; Rashid and Boussabiane 2019) have studied the connection between decision-makers' personal variables and the decisional output. In line with this literature, Wu et al. (2017) stated that the personal features of managers together with characteristics of their working environment could influence the decision-making process and operating choice. According to the author, the main decision-maker influencers are: (i) the manager's international experience; (ii) the specific task/role; (iii) the number of team members with which the manager works; (iv) the working atmosphere and pressure. In other words, Wu et al. (2017) recognise that working experience/role and working context play a crucial role in influencing business decisions. Nevertheless, within the healthcare managemen<sup>t</sup> field, a literature gap was also noted about the specific analyses of the linkage between intrinsic human variables of clinicians and the decisional biases connected to the clinical reasoning. Actually, this second aspect still remains understudied.

This is in accordance with what Ashoorion et al. (2012) have stated. By analysing the correlation between physicians' personal variables and clinical reasoning on medical students, authors have claimed the need for future studies to confirm their results in other fields. With regards to the working context, only a study by Elvén et al. (2019) has defined those variables as able to e ffectively influence clinical reasoning. Nevertheless, their sample study has referred only to the physical therapist students and not to physicians; thus, also this aspect would deserve to be further investigated.

Accordingly, this article would like to contribute to filling the two above-mentioned gaps arisen from the literature, with a specific reference to the decision-making process in the surgical field. To achieve this goal, the inquiry intends to analyse both features of decisions in the orthopaedic context and physicians' personal variables which would trigger cognitive biases and, in turn, mistakes in clinical choices.

Starting from this precondition, this study wants to understand which kind of errors are made by orthopaedic surgeons during their clinical decision process related to the follow-up of hip and knee arthroplasty. As a secondary endpoint, consequent to the main one, this study intends to verify the e ffective contribution of cognitive tools in recognising biases in the orthopaedic field, whose investigation would improve the quality of clinical decision-making path.

For the sake of clarity, the orthopaedic surgical field is chosen due to its standardised process and its considerable impact on healthcare systems' expenses. Mainly, the choice of the hip and knee arthroplasty (the most common joint reconstructions) is exactly due to the high frequency and demand that characterise the two surgery procedures (Ministero della Salute 2016). In fact, as reported by Bcc Research (2020) estimates, the world market relating to joint reconstructions should reach a value of 26.81 billion dollars by 2025, with a number of interventions equal to 5198.38 million. These two procedures, for example, represent the largest expense of implant costs in the USA. (Robinson et al. 2012).

Actually, the goal of a healthcare institution is creating higher value for patients at lower costs (Porter 2009); thus, the availability of information about the e ffectiveness and e fficiency of the treatments at every stage of the patient value-chain (from hospital admission to follow-up) becomes fundamental. In order to address this challenge, managers of healthcare organisations need solutions that would allow them to improve decision-making and business processes together with communication among doctors, patients, and administration, as well as e ffective access to di fferent data (Olszak and Batko 2012).

For this study aim, however, it is fundamental to specify the concept of follow-up to which we refer: "a check on someone who has been examined before in order to assess the process of a disease or the results of treatment" as defined by Dictionary of medical terms (Collin 2009). The moment of the follow-up decision was choosing to analyse a different aspect of clinical decision-making. Most studies tended to focus on clinical decision making within the hospital and in an emergency situation (Flynn et al. 2012; Lo and Katz 2005; Hess et al. 2015); on the contrary, we want to analyse the process related to the long term managemen<sup>t</sup> of the patients and clinical decision-making process that condition the continuum of care (Jette et al. 2003) in hip and knee arthroplasty.

To sum up, the aim of the study, in relation to the decision-making process of orthopaedic surgeons about the follow-up of hip and knee arthroplasty, is twofold:


To achieve these goals, the work follows three main stages. First of all, an analysis of the theoretical background on decision-making and its criticalities and connections with the healthcare and surgical fields was carried out. Second, it was run an interview-based qualitative case study on the field of orthopaedic surgery; the interviewees are well-informed physicians representing a good cross-section of the Italian orthopaedic landscape.

Lastly, besides verifying those consolidated biases in healthcare decision-making, the analysis of qualitative findings demonstrates a connection between some specific biases and two features of decision-makers: working experience and working context. This last issue, particularly, represents an aspect of healthcare decision-making still understudied; this would pose the study as a novelty in the field.

The paper proceeds with the following outline: after this introduction, the second section focuses on the main theoretical background. The third section reports the methodology of the study, while the fourth one presents the research findings. Section five discusses the results obtained and provides some considerations about the endpoints of this study. The last section includes the final remarks.
