Patient involvement and language barriers: Problems of agreement or understanding?

Highlights

• Patients minimal responses can indicate problems of agreement or understanding.

• Patient-centered behavior can be deficient in difficult decision-making.

• Patient-centered behavior may not secure understanding and elicit patients views.

• Interactional analysis can identify strategies that impede patient participation.

• Interactional analysis can identify strategies that enhance patient participation.

Abstract

Objective

This study aims to explicate efforts for realizing patient-centeredness (PCC) and involvement (SDM) in a difficult decision-making situation. It investigates what communicative strategies a physician used and the immediate, observable consequences for patient participation.

Methods

From a corpus of videotaped hospital encounters, one case in which the physician and patient used Norwegian as lingua franca was selected for analysis using conversation analysis (CA). Secondary data were measures of PCC and SDM.

Results

Though the physician did extensive interactional work to secure the patient’s understanding and acceptance of a treatment recommendation, his persistent attempts did not succeed in generating the patient’s participation. In ratings of PCC and SDM, this case scored well above average.

Conclusion

Despite the fact that this encounter displays some of the ‘best actual practice’ of PCC and SDM within the corpus, our analysis of the interaction shows why the strategies were insufficient in the context of a language barrier and possible disagreement.

Practice implications

When facing problems of understanding, agreement and participation in treatment decision-making, relatively good patient centered skills may not suffice. Knowledge about the interactional realization of key activities is needed for developing training targeted at overcoming such challenges.

Introduction

Patient-centered care (PCC) has become the norm of quality health care in many countries. A central component in PCC is the involvement of patients in treatment decision-making, i.e. through exploring patients’ preferences and concerns, as conceptualized in shared decision making models (SDM) [1], [2], [3]. Patients’ opportunities to accept or reject treatment recommendations can be seen as a basic form of patient involvement, built on the ethical requirement of informed consent [4], [5], [6]. A precondition for accepting or rejecting a proposal is to understand it [7], [8], and most PCC and SDM guidelines advise physicians to check and clarify understanding [1], [3], [5]. However, patients do not necessarily express their lack of understanding, or they may overestimate what they have understood, whereas physicians may overestimate the clarity of their own talk, and rarely check what patients have actually understood [9], [10], [11], [12].

Achieving and securing understanding may be particularly challenging in encounters with non-native speakers [13], and ineffective communication with non-native speaking patients constitutes a risk to patient safety and health [14], [15], [16]. Interpreters can be used to overcome language, culture and knowledge barriers [17], but interpreters are not always used when patients have ‘some’ proficiency in the second language [18]. Despite these challenges, little is known about what communicative strategies physicians actually use to secure understanding on a micro level, turn-by-turn, in authentic monolingual and multilingual encounters [13], [19], let alone how understanding is accomplished in situations where not only the patient, but also the health professional speaks a non-native language. The use of lingua franca is far from uncommon in contemporary multilingual societies, where immigrants partake considerably in the health care work force [20], [21]. Contributing to fill this gap in research, the present study takes a conversation analytic approach in order to explore interactional strategies and consequences in a decision-making sequence where the physician and patient, with different native languages, use Norwegian as a lingua franca. The analytical starting point was to investigate the physician’s various attempts to secure understanding and generate participation from a seemingly ‘passive’ patient, whose dominant contribution was minimal responses.

Minimal responses, such as “mm” and “yes,” serve a variety of functions in talk. Which function is realized in a given instance depends on such things as prosodic delivery and both what it is responsive to and what happens next, making them a useful, but possibly ambiguous resource for communication. Minimal response tokens claim understanding, by passing the opportunity to initiate repair and giving a go on-signal to the speaker, but do not display any evidence of what has been understood [22], [23]. Thus, minimal responses provide weak evidence of what is actually understood. Indeed, a study of simulated physician–patient interaction found that, in multilingual dyads, minimal responses were misleading in terms of recipient recall [24]. Another experimental study demonstrated that minimal responses, produced by distracted listeners, in positions where more specific responses (i.e. assessments) would be expected, affected the quality of speakers’ narratives negatively in that speakers, for instance, “circled around and retold the ending more than once” [25].

The impact of minimal responses has been amply demonstrated in decision-making sequences across various settings. A study of ordinary conversations showed that proposals for future action require explicit statements of commitment and not merely a minimal confirmation in order for the proposal to be heard as accepted by the addressee [26]. A similar pattern has been found in treatment recommendation sequences, where physicians regularly treat acceptance of their treatment recommendation as necessary before moving on to the next activity [27], [28], [29]. In most settings, physicians treat patients’ minimal responses, like “mm”, as insufficient acceptance (i.e. displaying passive resistance), while explicit or elaborate affirmative responses (e.g. “okay”, “that sounds good”) are required to be heard as accepting the proposal [28], [29], [30]. When such acceptance is not forthcoming, a negotiation sequence usually follows, dealing with potential problems of acceptability. These two trajectories of treatment recommendation sequences are illustrated in Fig. 1.

The present study examines a series of treatment recommendation sequences that all fall under the second trajectory (see Fig. 1). The analysis focuses on the third part, on what the physician does after responses by the patient that are heard as insufficient acceptance. The study aims to explicate efforts for realizing patient centeredness in an encounter where achieving patient participation in decision-making was particularly challenging. It provides empirical specification of 1) what communicative strategies the physician used in order to overcome problems of establishing understanding and generating patient participation, and 2) what observable consequences the physician’s efforts had for the patient’s understanding and participation within the encounter.