Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

Face-saving strategies and the burden of opioid policy enactments:
When physicians’ compliance makes patients non-compliant
Peter Joseph Torres, PhD
Miami University, Oxford, Ohio, USA
Abstract
The escalation of the opioid epidemic in the United States has sparked sweeping
legislation meant to regulate physicians’ opioid prescribing practices. The demands of such
policies force physicians to initiate discussions that could jeopardize the collaborative doctorpatient relationships necessary for curbing inappropriate opioid prescriptions. Drawing on
sociopragmatics, this discourse analysis study of primary care interactions examines the facesaving linguistic features employed by physicians in negotiating the line between policy
demands and maintaining collaborative relationships. The findings reveal several face-saving
acts—pseudo requests, downtoners, broadening, redirection, tag questions, impersonalization,
listing, and (negative) imagery—used by physicians when enacting the three most prominent
policies: (1) monitoring opioid use, (2) prescribing anti-overdose medication, and (3)
transitioning patients from opioids to alternative treatment. Informed by Goffman’s concept of
“face-work,” this study provides evidence of the communicative burden placed on physicians
implementing disagreeable opioid policies, as well as opening up discussions on how
policymakers and medical institutions can support physicians in implementing opioid policies.
Keywords: opioids, face-work, face threats, medical discourse, doctor-patient interaction,
discourse analysis, sociopragmatics

‘The final, definitive version of this paper has been published in Journal of Pragmatics
Volume 205, February 2023, Pages 122-136 by Elsevier Publications Ltd,
https://doi.org/10.1016/j.pragma.2023.01.001 All rights reserved. © Peter Joseph Torres

1

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

1. Introduction
The opioid epidemic in the United States has prompted policymakers all over the country
to pass sweeping legislation addressing physicians’ prescribing practices (Al Achkar et al.,
2017). The crisis has also prompted numerous multidisciplinary investigations analyzing medical
interactions involving chronic pain and prescription opioids, which continually provide
overwhelming evidence that collaborative relationships—devoid of conflicts and
disagreements—facilitate treatment goal-setting and reduce inappropriate opioid prescribing
(Henry et al., 2016; Hood-Medland et al., 2021).
Despite ever-increasing interest in both areas, the relationship between policies and their
linguistic enactments has been overlooked; therefore, policymakers have missed valuable
opportunities to assess implementation at a discursive level and identify pressing issues in need
of attention. For instance, the demands laid out by policies force physicians to initiate potentially
tense exchanges about overdose and addiction, conceivably jeopardizing the collaborative
relationships necessary for effective pain management. Grounded in sociopragmatics—the study
of meaning expressed through language and its relationship to context (Sifianou & Tzanne,
2021)—this study uncovers the discursive strategies physicians use when discussing a topic as
fraught with difficulty and controversy as opioids. This study bridges policy with local
interactions, using discourse analysis (DA) to outline the face-saving practices physicians
employ while carrying out increasingly restrictive opioid policies.
The analysis of this study is also couched within Goffman's (1955) concept of facework—that is, the constant negotiation, construction, and protection of one’s desired social
persona, as well as that of their listeners. Through the discourse analysis of audio-recorded
primary care visits, this study seeks to answer the research question What face-saving discourse

2

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

features or practices do physicians employ when enacting opioid policies? The investigation
found three different policies that consistently came into play during these interactions, within
which were eight salient face-saving discourse features employed by participating physicians.
Recognizing the challenging aspects of opioid-related discourse foregrounds policy
issues that need addressing, opening up discussions for how policymakers and medical
institutions can support physicians in enacting opioid policies. Moreover, examining how
physicians use language to interpret and enact opioid policies provides a sociopragmatic
understanding of the dilemmas physicians face, as they know very well that their adherence to
restrictive opioid policies may result in the non-compliance of patients. This study locates itself
within the burgeoning body of work interested in using discourse analytic approaches to
uncovering the pragmatics and meaning-making process involved in local health policy
enactments (Martínez, 2008; Ramanathan, 2010; Evans-Agnew et al., 2016).
2. Related Literature
The stigma associated with discussing opioids and addiction has made healthcare
communication surrounding pain fraught with difficulties (Arborelius & Thakker, 1995) for both
providers (Matthias et al., 2013) and patients (Torres et al., 2020; Roberts & Kramer, 2014;
Henry et al., 2017). Patients taking opioids have claimed that they were made to feel as if they
were “drug addicts,” “junkies,” and “heroin users” (Dassieu et al., 2021, p. 5); meanwhile,
physicians found opioid discussions to be equally complex, citing reasons such as fear of
resistant behavior (Mistral & Velleman, 2001), distrust of patients who initiate opioid-related
conversations (Hughes et al., 2015), doubts surrounding patients’ willingness to disclose
substance dependence or abuse (van Boekel et al., 2013), and lack of training (Klamen, 1999).
Such a highly charged issue as opioids pose a challenge to physicians who have to negotiate
3

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

between enforcing restrictive policies patients find disagreeable and maintaining collaborative
relationships integral to treating chronic pain.
2.1. Face, face threats, and face-work
Erving Goffman (1955, p. 213) is the pioneering theorist who conceptualized “face” as an
individual’s desired “positive social value” derived from “approved social attributes.” Goffman,
whose research in psychiatric wards exemplifies the shared history between face and medical
interactions, described the harmful or inaccurate portrayal of one’s face as “threats” (p. 217) and
the actions one takes to maintain or save face as “face-work” (p. 216). Brown and Levinson’s
([1978]1987, p. 61) elaboration of Goffman’s thesis bridges the notions of face and politeness,
differentiating “positive face” (the “desire to be appreciated”) from “negative face” (the “desire
to be free from imposition”) as motivations for “politeness”—the mitigation, avoidance,
disarming, and presupposing of “face-threatening acts” (FTAs).
Although Brown and Levinson’s contributions served as a sounding board for subsequent
face research, their work received criticism for its incompatibility in cross-linguistic discourses
(Matsumoto, 1998; Terkourafi, 2012) and perceived inattention to the impact social identity has
on self-image (Bogdanowska-Jakubowska, 2010, p. 217). O’Driscoll (2017, p. 90) also
questioned Brown and Levinson’s sometimes interchangeable, sometimes causal treatment of
face and politeness, claiming that face is a trait speakers own while politeness is a behavior
speakers perform. Spencer-Oatey's (2008, p. 12) "rapport management" framework sought to
address these issues by incorporating context-specific goals and expectations, along with face, as
necessary components for maintaining harmonious relationships. Locher and Watts’ (2005, p.
10-12) also proposed the alternative “relational work" framework, which describes one's efforts
to create or maintain a socially mediated face. Although theoretically distinct, the overwhelming
4

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

similarities between rapport management and relational work have led to studies describing both
as "virtually synonymous with face-work in its widest sense" (O’Driscoll, 2017, p. 101; see also
Sifianou and Tzanne, 2021, p. 259), but without the association to Brown and Levinson's
individual-centric take.
While there are considerable disagreements concerning the relationship between facework and politeness, there seems to be consensus acknowledging these concepts as contextual
(Culpeper, 2011), constantly negotiated (Ting-Toomey and Kurogi, 1998), and distinct from one
another (Watts, 2003). Thus, this study broadly situates itself at the center of these parallels,
adopting Spencer-Oatey’s (2008, p. 14) more holistic take on face as an individual’s “sense of
worth, dignity, and identity, associated with one’s status, reputation, and competence.” SpencerOatey (2008, p. 21) considers face management vital to maintaining harmonious relationships
during face-threatening discourses, such as "raising sensitive topics." This study also adheres to
Locher and Watts’ (2005, p. 10–12) view that face is constructed discursively through
interactions between individuals engaged in relationship negotiation, as opposed to Brown and
Levinson’s ([1978]1987) position in which face inherently resides in the individual. This
interpersonal view of face is in keeping with this study’s goal of identifying face-work strategies,
whether or not their subjective functions are to express politeness (good manners).
2.2. Face-work in medical discourse
The very nature of the occupation pushes physicians to initiate face-threatening
discussions, from asking value-laden and sensitive questions to providing potentially
demoralizing evaluations (Benkendorf et al., 2001). Aronsson and Sätterlund-Larsson (1987)
even referred to medical interactions as a social choreography where patients and physicians
constantly dance around sensitive discussions through tactics like embedding requests into
5

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

suggestions, hedging through the use of modal verbs, and providing preemptive apologies.
Ultimately, they argue that such a dance begets negative consequences on the joint decisionmaking of patients and physicians.
Various discourse features used by physicians in discussing similarly charged topics as
opioids have been identified in the literature. Lutfey and Maynard (1998) highlighted
oncologists’ use of euphemisms and avoidance of the words “death” and “dying” when breaking
the news to cancer patients that they are no longer treatable. Similarly, Wilkinson and
Kitzinger’s (2000) observation of focus group meetings found the redirection of cancer talk
toward positive thinking and the use of words such as “positive” and “positively” as discursive
coping mechanisms used by female patients. Meanwhile, Epstein et al. (1998) found evidence of
incoherence and fractured speech in physician speech when discussing HIV diagnoses. Caffi’s
(1999) analysis of doctor-patient transcripts have called attention to physician’s use of
downtoners such as “just” or “by any chance” in asking difficult questions, and the inclusive we
to lessen social distance and cushion the weight of mandates. Lastly, Clark and Hudak’s (2011)
study of orthopedic surgeons revealed that the use of (1) tag questions like “You know?”
positions patients as complicit to their recommendations; (2) redirections to standard practice
help mitigate the intensity of treatment plans, and (3) general case descriptions (referred to here
as broadening) or the referencing of patient’s group membership (e.g., "younger patients" or "lots
of folks") as justification for treatment suggestions.
The current study adds to the existing literature that pays attention to the language of
medical discussions surrounding sensitive topics, underscoring the importance of recognizing
and addressing face threats as a critical first step in improving doctor-patient discourse.

6

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

2.3. Opioid policies
This study uses Torres’ (2021, p. 2) linguistically viable definition of policy: “chunks of
language (discourse) made up of lexical and grammatical features that denote a suggestive intent
of regulatory measures and courses of action concerning a given issue. Torres (2021) showed
how the language used in framing opioid policies has become increasingly restrictive as the
epidemic worsens in the United States. As stricter policies surrounding opioid prescriptions are
implemented, discussions about opioids are becoming more tense, both for physicians who are
under heightened scrutiny and for patients whose opioid treatments are changed or requests
rejected (Henry et al., 2016).
By calling attention to specific face-saving features physicians employ when complying
with new health directives, this study contributes a physician-centered analysis to the existing
literature on face-work, which has until now primarily focused on patient discourses (AinsworthVaughn, 1998). Although this work focuses solely on primary care consultations in California,
studies on policy-driven face-work and politeness routines are not unique to the medical space
and can be replicated for almost all workplaces and institutions where policies exist.
3. Methodology
This study analyzes the audio recordings of doctor-patient interactions at the City
Medical Center (CMC; a pseudonym). CMC is an ideal setting because the organization
considers balancing pain treatment and drug diversion one of its top priorities.
The logistics involved in recruitment and data handling are compliant with the criteria
outlined in the Health Insurance Portability and Accountability Act (HIPAA) and were
conducted under the guidance and approval of CMC’s Institutional Review Board. Recruitment
started by first enrolling attending and resident physicians into the study to ensure that only the
7

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

patients of participating physicians were recruited for this project. In residency clinics and
teaching hospitals, residents are medical school graduates who are continuing their specialty or
general practice training under the supervision of a more experienced attending or faculty
physician. A total of 30 physicians provided consent and enrolled in the study. The next step was
to recruit the enrolled physicians’ patients who receive prescription opioids for chronic pain.
Patients receiving opioids for cancer pain were excluded since the policies and the context in
which they use opioids are different from those of chronic pain patients.
Patients were recruited in waiting rooms after they checked in with the front desk. Once
the rooming process of a consenting patient was finished, the researcher entered the examination
room to check in with that patient, answer any final questions they may have about the study,
and set up the recording device. The recorder was positioned away from the patient’s line of
sight to avoid causing distractions. Then, the researcher stepped out of the room, leaving only the
patient and their companion (if present) in the room to wait for the physician. The researcher was
not in the room during the recording to avoid any discomfort by the participants. At the end of
the visit, the researcher returned to the room to end the recording, retrieve the recorder, and
administer a post-appointment survey eliciting relevant information on the patient’s linguistic
background. The average appointment duration was one hour.
Ultimately, three patients who fit the criteria agreed to participate in this study. Table 1
outlines relevant self-identified information provided by the participants in their post-visit
questionnaires:
Table 1: Participants’ self-identified information
Resident physician

Patient

Pseudonym

Age

Pseudonym

Age

Dr A
Dr B

29
30

Pt A
Pt B

40
52

8

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

Dr C

28

Pt C

69

Note: In the excerpts presented, Com refers to the patient’s companion while Att refers to the attending or faculty physician
overseeing the resident.

The choice to look at a small number of participants is appropriate in investigations
involving language-in-action, as it requires and allows for a thorough analysis of the discourse
contexts and features in question (Schilling-Estes, 1998). All participants expressed that English
is their native and preferred language in medical appointments. All physicians were considerably
younger than their patients, making age differences consistent enough to potentially mitigate
some age-related or generational variation. Later in this study, I advocate for future explorations
to emphasize sociolinguistic variations among and within diverse groups—from race groups and
self-identified gender to education and social status, among many others—as more data become
available. The outcomes of this study are not intended to be descriptive of all opioid-related
interactions; rather, the findings are reported based on their potential to inform future interactions
of a similar nature and to highlight the challenges institutions impose upon policy stakeholders
when not enough attention is given to the language required to implement potentially contentious
policy mandates in medical interactions.
Transcriptions were organized into speaking turns—all utterances of a speaker before
another mediates (Sacks, Schegloff, & Jefferson, 1974)—instead of individual utterances
because physicians enacted one policy at a time, which means all utterances within a single turn
covered the same policy. Aside from efficiency, speaking turns allows for a widely accessible
representation of data in this study. Transcripts were prepared by the researcher with the help of
three research assistants trained in discourse analysis and certified by the Collaborative
Institutional Training Initiative. Each transcriber worked on the recordings separately before
coming together to settle discrepancies, adding another layer of reliability check.

9

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

Segments of sample excerpts shown here, including suprasegmental features, were
trimmed for readability and brevity in respect of the publication's length limitations. Care has
been taken to guarantee that the omissions, which are indicated by three dots "...", have no
bearing on the features presented. Unabridged transcripts can be found in Torres (2022b) or by
request from the author.
3.1. Policies
Three newly chaptered state policies, detailed in Table 2, consistently emerged from the
recordings. These policies were also incorporated in the agreement both patient and physician
must sign before initiating an opioid regimen. (Brown Jr et al., 2014).
Table 2: State opioid policies analyzed in this study
New policy

Description

1. Continual monitoring/
heightened surveillance
of opioid use.

Physicians must subject patients to routine drug tests to
ensure adherence to opioid regimen.

2. Co-prescribing antioverdose medication with
opioids.

Physicians must prescribe opioid antagonists like naloxone (e.g., brand
name Narcan) alongside opioids.

3. Transitioning to
alternative treatment/
Weaning patients off
opioids.

Physicians must discuss alternatives to narcotics, such as non-steroidal
anti-inflammatory drugs (e.g., ibuprofen), acupuncture, and other
procedures (e.g., trigger point injections). Physicians must conduct
additional tests to identify potential source of pain (e.g., MRI).

Note. These policies can be found in Brown Jr et al., (2014).

3.2. Analytic approach
Discourse analysis—the study of language relative to the context in which it is used (Gee,
2011)—was chosen as primary analytic approach because the research question calls for the
understanding of interlocutors’ meaning-making processes beyond the syntactic level; such an
understanding includes the motivations behind eliciting certain pragmatic practices and the
communicative means through which these discursive agendas are realized (Kampf, 2015, p.3;
10

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

Johnstone, 2018). As Woods (2004) explained, DA allows researchers to make sense of the
motivations behind the language choices of healthcare providers. Moreover, DA captures the
complexity of medical interactions in a way that can be made accessible to a general audience,
including healthcare providers, without watering down the substance.
The specific analytic approach used in this study is essentially the same as Benkendorf et
al.’s (2001, p. 202) “data-driven approach” and Roberts and Sarangi’s (2005, p. 632) “themeoriented approach” to studying medical encounters. With the help of the recordings and
transcripts, instances in which the physicians used face-work to enact opioid policies were
identified. The process required assessing the range of functions served by each turn relative to
the context of the ongoing exchange. In other words, identifying face-work meant making sense
of the linguistic choices made by physicians. This strategy is derived from Van Dijk’s (1999)
context model framework—a schema designed to reduce the complexity of social situations by
probing language relative to the available contextual information (see Torres, 2022a). The
identified features were systematically classified as connections and similarities between them
became apparent. This way, the data dictates the categories, not forced into pre-identified ones.
As is typical in discourse analysis, findings could either be novel or features prior studies have
already found in other contexts (e.g., discourse markers, downtoners, and broadening, among
others discussed in Section 2) and are, thus, identified as such.
4. Results and Discussion
Eight face-saving features emerged from the physicians’ speech as they enacted the three
CMC policies outlined in Section 3.1. These findings are summarized in Table 3.
Table 3: Face-saving acts used by physicians in enacting CMC’s opioid policies

11

✓

Co-prescribing naloxone

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

Transition to alternative treatment

(Negative) Imagery

Listing

Impersonalization

Tag questions

Redirection

Broadening

Downtoners

Pseudo requests

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

New CMC opioid policies
Continual monitoring/Drug testing

✓

✓

In the following subsections, I go through each strategy and provide excerpts
exemplifying their use in the enactment of opioid policies.
4.1. Pseudo requests
Participating physicians used pseudo requests—imperatives and mandates presented as
requests and suggestions instead of direct, declarative commands (Benkendorf et al., 2001)—to
save face when subjecting their patients to drug tests. Excerpts A and B show examples of
pseudo requests, indicated by the bolded text:
(A)
A1 Dr. A: Okay. They’re also going to give you the flu shot, and we’ll have to do a urine
test. Is that okay with you?
A2 Pt. A: Uh huh.
[Physician leaves then re-enters the room.]
A3 Dr. A: Alright, can you get your lab checked real quick upstairs?
A4 Pt. A: No, I’m already runnin’ too late.
(B)

B1 Dr. B: Okay, my instruction for you today is that maybe you have time to go upstairs to
do some quick labs?
B2 Pt. B: [Laugh] [inaudible] Why don’t you just keep me here?
12

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

B3 Dr. B: [Laugh]... It should be really quick. I know it’s been a lot for you today. It’s just,
you know, it’s just, it’s our policy and it’s a government mandate to do these labs if
you’re getting prescribed with narcotics. It’s all for your safety.
B4 Pt. B [inaudible] killing me. No. [laugh] I’m joking.
B5 Dr. B: It’s, it’s just rules, you know? And I know you’ve been a good patient. It’s just
routine unfortunately. In fact, it’s in our contract, remember?
In both A1 and B1, the physicians' enactment of the drug testing policy started in the
declarative form yet ended as questions. Dr. A's sudden pivot stood out because it ensued
immediately after their mention of flu shots and urine test as near collocates, evoking some sort
of pragmatic equivalency between urine toxicology screenings and the more mundane seasonal
flu shots. On top of that, their use of the modal phrase we('ll) have to do added some sense of
necessity to the drug test. Similarly, Dr. B's request seemed non-negotiable at first before hinting
at some tentativeness with the word maybe. As Dr. B said in B5, being subject to regular drug
screening is incorporated in the agreement patients sign before starting an opioid regimen.
However, seeking the patient's approval and posing the mandate as a question mitigates the
intensity of the command. This observation is in line with Benkendorf et al.’s (2001) study of
counseling discourse, which shows that genetic counselors use pseudo requests to politely
facilitate their clients’ decision-making, making it seem as if they arrived at a decision together.
While treating patients as individuals with agency could help foster collaborative relationships,
framing mandates as requests makes them vulnerable to rejection, as seen in A4, or pushbacks,
like in B2 and B4. This policy provides a clear example of the irony between physician
compliance resulting in patient’s non-compliance.
Another noteworthy quality of the pseudo requests found in the data was that they cooccurred with discourse markers such as okay and umm in Lines A1 and B1. Beach (1993; 2020)
wrote extensively about oncologists’ use of discourse markers like okay as transition devices that

13

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

delicately end patients’ extended turns while simultaneously signaling alignment as patients feel
heard.
4.2. Broadening
I use the term “broadening” to describe the three strategies or features employed by
physicians to widen the scope of ongoing discussions: (1) Audience generalization, (2) inclusive
pronouns, and (3) modal verbs.
4.2.1. Audience generalization
I refer to the process of generalizing the intended recipient of policy demands as audience
generalization. We have already encountered this process in B3 when Dr. B clarified that the
drug test is not an isolated event, rather, it is a policy that applies to Pt. B because of their group
membership—as individuals receiving prescription opioids. Physicians also used audience
generalization when co-prescribing the anti-overdose drug, naloxone to preempt possibilities of
patients feeling targeted or misjudged. Take, for example, Excerpt C below. The bolded text
represents broadening.
(C)
C1 Dr. B: Do you have a question about the naloxone or the Narcan that we’re offering to
every patient?
C2 Com: What is that?
C3 Dr. B: It’s a medication, a nasal spray, that reverses an opioid overdose... everyone gets it
since we are experiencing an opioid crisis.
C4 Pt. B: [laughing] I don’t [inaudible] any medication [inaudible] as a matter of fact.
C5 Dr. B: Yeah, yeah, no, I get that ... we’re offering it to all our patients that get, um,
opioid prescriptions.

In C1, rather than volunteering information, Dr. B. solicited from Pt. A, a request for
additional information about naloxone. The strategy stood successful, as demonstrated by the
companion’s question in C2, ultimately leading to Pt. B introducing the overdose medication as a
14

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

question response in C3. The laughter and defensive reaction in C4—when Pt. B expressed not
needing such medication as a matter of fact—suggests decisive disagreement. However, the
physician saved face by constantly referencing a broader audience.
4.2.2. Inclusive pronouns
The physicians also used inclusive pronouns like we and our to frame policy enactments
as a joint or collective effort. Take, for example, Excerpt D:
(D)
D1 Dr. C: I need you to do some blood work
D2 Pt. C: Okay. I will.
D3 Dr. C: One of them is for Norco. We just need to do that [drug test] every so often.

Here, Pt. C already agreed to completing the blood work, yet Dr. C felt the need to
specify that one of them is for screening opioids. In D3, Dr. C further softens the endeavor
imposed on the patient by stating that the test is not done often, coupled with the inclusive we,
which contrasts the pronoun I in D1. Inclusive pronouns mainly occurred during physicians’
enactment of the policies on drug screening and suggesting alternative treatment. One way to test
the pragmatic weight of the features discussed in this study is to assess the alternatives. Figure 1
contrasts the physicians’ use of inclusive pronouns (shaded) against “non-inclusive” alternatives
(clear)

15

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

Figure 1: Physician utterances with and without inclusive pronouns

The contrast highlighted in Figure 1 illustrates physicians’ use of inclusive pronouns to
signal, if not solicit, alignment, consistent with earlier studies describing their effectiveness in
reducing social distance (Caffi, 1999). This study contributes a policy perspective to the
discussion, demonstrating the use of we and our as face-saving tools that reframe policy enactors
as stakeholders or recipients subjected to policies.
4.2.3. Modal verbs

The last broadening tool physicians used was modality. The use of modal verbs in
medical discourse has been shown to suggest physicians’ willingness to negotiate (Brown and
Levinson([1978]1987). Modals have also been shown to convey optionality in policies because
their permissive and restrictive qualities allow stakeholders to reevaluate what would otherwise
be straightforward directives (Torres, 2021). Figure 2 shows just that — the absence of modals
renders unmitigated commands. The shaded segments show the physicians’ use of modals, while
the clear segments present modal-less alternatives (clear).
16

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

Figure 2: Physician utterances with and without modal verbs

Note. For the alternative versions, the second-person pronouns can also be omitted with the modal verb, as what usually happens
when their referent is apparent to the interlocutors.

As shown in the first two examples in Fig. 2, adding modals in commands conveys
malleability, leaving them susceptible to rejection. Modals have the same effect on statements.
For instance, saying it should be really quick (in Fig. 2) allows patients to consider instances in
which drug tests are not quick, as opposed to it is really quick, which evokes a higher degree of
certainty. These findings show that modals function as a broadening tool in verbal policy
enactments just as much as what has been shown in written health policies (Torres, 2021).

17

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

4.3. Impersonalization
This study adopts Rundblad’s (2007) description of impersonalization—a rhetoric device
used in scientific writing to frame subjects as if they could have been anyone else—into the
context of face-work in spoken discourse. Impersonalization is often associated with agentless
passive constructions in written communication (Brown & Levinson, [1978]1987, p. 273), but,
as this study will show, impersonalization emerges in myriad other ways—from repairs to
wordplay—in spoken interactions. The bolded text in Excerpt E are examples of
impersonalization:
(E)
E1 Dr. A: Yeah so, we also gave you naloxone. This is an injection that you take—
E2 Pt. A: —I got a whole bunch of boxes of that—
E3 Dr. A: ...if you, if, if a, if there’s concern for overdosing, on any of the other ... like
Norco or even like, you know, other substances like heroin or anything like that,
other things that may, you know? It’s still important to have this just in case.
E4 Att: Yeah, it’s just required by the law.

Because common knowledge of healthcare transactions suggests that physicians prescribe
the medication one needs, prescribing naloxone may be perceived by patients as insinuating that
they misuse prescription opioids or heroin. To avoid coming across as accusatory, Dr. A avoids
mentioning any definite human agent (i.e., one who overdoses) or experiencer (i.e., one who
suffers an overdose) in E3. Dr. A’s first mention of you in E3 may have initially positioned Pt. A
as the agent. Still, the physician immediately engaged in a repair—the altering of meaning or
direction of ongoing speech—to obfuscate any identifiable agent, as shown in the ensuing
syllabic repetition: if, if a, if. The motivation to impersonalize demonstrates just how complicated
it is to implement opioid policies.

18

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

The following example shows a segment of Dr. A suggesting alternative treatment.
Specifically, Dr. A was discussing the harmful effects of opioids to convince Pt. A into trying
alternative remedies.
Dr. A: you can come to the hospital cause you—some people can get, um, overdose on
the opiates. there’s a risk of sudden death

Here, the physician engaged in a repair by immediately inserting the phrase some
people after uttering the pronoun you. Then, Dr. A continued to omit identifiable agents while
discussing overdoses: there's a risk of sudden death as well. Without impersonalizing through
repair and agent deletion, the ensuing phrase would have been You're at risk of sudden death
More examples of impersonalization are found in (F), represented by the bolded text.
(F) Additional examples of impersonalization:
F1 Dr. B: [naloxone] reverses an opioid overdose. It also works for heroin overdoses ...
F2 Dr. B: ... It’s just that no one really plans for an overdose, you know?
F3 Dr. B: In fact, what I tell people, you know, is that you could, you know, help rescue
someone who’s had an opioid overdose at home or even heroin overdose in the
streets in your community.

In F1, Dr. B avoided using an agent as they volunteered the information that Narcan is
just as effective in reversing overdoses from heroin as it is with prescription opioids. Patients
may wonder why their physician, who knows they already take opioids for pain, would bring up
heroin. As much as the discussion was already face-threatening, it could be worse if Dr. B had
explicitly associated Pt.2 with heroin. Other examples of impersonalization would be Dr. B's use
of no one in F2 and someone in F3. F3 is unique because it shows Dr. B employing both
impersonalization in the context of overdosing and its opposite (i.e., framing the subject as an
active participant) in the context of helping others. This seemingly simple yet innately

19

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

complicated construction allowed Dr. B to frame Pt. B as someone who could save rather than be
saved.
4.4. Redirection
Another strategy physicians use to mitigate looming opioid discussions is to divert the
conversation toward alternative, less-threatening paths. For instance, physicians broached
tapering discussions by redirecting their patients’ attention toward opioid side effects—a list too
long that it gives physicians plenty of opportunities to initiate tapering discussions. For example,
Excerpt G shows a physician addressing their patient’s complaint about itchiness by citing
opioids as its likely cause in order to initiate tapering discussions.
(G)
G1 Dr. C: The one thing I could think of on your list that would make you itchy is your Norco.
G2 Pt. C: No, I don’t think so cause I had the Norco after I started itching.
Here, Dr. C used redirection to shift the discussion from the patient’s complaint about
itchiness to opioid side effects. Using the lengthy list of opioid side effects as the reason for other
medical issues provides physicians plenty of opportunities to discuss alternative pain treatment.
The existing literature on medical communication involving opioids has hinted at the use of
redirection in enacting policies. For instance, Henry et al.’s (2019) research recognized situations
in which patients presented negative evaluations of opioids or made mention of their side effects
as promising opportunities for physicians to broach the subject of lowering opioid doses or
starting non-opioid treatments. The same approach is shown in G1, except it was the physician,
not the patient, who gave a negative criticism of opioids in an attempt to redirect the
conversation toward alternative treatment. Another way physicians initiated the discussion of

20

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

tapering is by redirecting their patient’s attention toward addressing the actual source of pain,
which is shown in Excerpt H.
(H)
H1 Dr. A: Um, and we kind of discussed maybe pursuing an MRI of your low back to see if
maybe there might be some pain related to that low back area and the spine.
H2 Pt. A: Well, I haven’t had nothing done to my—
H3 Dr. A: Yeah, but sometimes if you have bilateral pain in your legs, that could be from the
nerves in your back.
H4 Pt. A: But this pain [inaudible] for so long?
H5 Dr. A: Yeah. I also was wondering if maybe, you might have what we kind of talked about
last time, like opiate, opiate use disorder. Basically your brain gets tolerant to a high
dose of opiates, and then, in order to control that pain, you need more opiates, but
it’s kind of a downward spiral where your brain gets used to being on more —
H6 Pt. A: — No , No. It’s like damned if I do and damned if I don’t.
H7 Dr. A: Yeah, yeah, yeah.

In H1 and H3, the physician attempted to shift the ongoing discussion toward pursuing an
MRI to narrow down the potential source of Pt. A’s pain. When met with contention, Dr. A tried
redirecting the patient’s attention toward addressing opioid use disorder, suggesting tolerance as
a contributing factor to her pain and need for more opioids. The conversation ensues with a
defeated response from Pt. A in H6, It’s like damned if I do and damned if I don’t. Dr. A’s vague
and seemingly defeated response in H7 is indicative of how difficult conversations about opioids
can be. The physician’s use of euphemism through medical terminologies such as opiate use
disorder is a face-saving feature that has already been extensively discussed in face-work
literature and, thus, is not dealt with in this study.
Physicians also employed redirection to minimize their responsibility in mandating drug
tests and co-prescribing naloxone, using phrases such as: It’s our policy and it’s a government
mandate (in B3), It’s just rules you know. It’s just routine unfortunately. In fact, it’s in our
contract, remember? (in B5), Yeah, it’s just required by the law (in E4).
21

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

Circling back to Brown and Levinson’s ([1978]1987, p. 61) differentiation between
positive and negative face, redirection allows physicians to save positive face by preempting
blame in addition to also protecting the patients’ negative face by providing them with an
alternative entity on which they can take out their frustrations. Redirection has been found to be
a discourse feature therapists use when responding to blame (Friedlander et al., 2000). The term
redirection also fits with what Clark and Hudak (2011, p. 386) describes as the act of
formulating treatment decisions as if they are “products of impersonal logic involving standard
practice or what one tends to do.”
Lastly, it is also worth pointing out that physicians also embellished their redirection with
lexical items that signal alignment with the patient. For instance, the word unfortunately in “It’s
just routine unfortunately” allows physicians to convey empathy and solidarity over some illfated experience. By conveying that they share their patient’s view and recognize their
predicament, physicians are able to position their stance closer to patients and further from
policies.
4.5. (Negative) imagery and listing
The participating physicians broached the face-threatening discussion of reducing their
patients' opioid dosage by organizing their justifications into lists. According to Jefferson (1990),
lists are discourse tools that, when used in speech, render convincing messages because they
signal thoroughness, completion, and comprehensiveness. Jefferson (1990) introduced the idea
of "three-part lists" after observing that lists are often comprised of three separate entities. DoriHacohen (2020, p. 307) expanded on this idea by proposing using "long lists" with more than
three entities to sound even more compelling. To be even more convincing, physicians also
peppered such lists with words that could evoke undesirable images. These lexical choices form
22

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

what Tannen (2007) refers to as imagery, a discourse tool that creates involvement and shapes
imagination. In Excerpt I, Dr. A used listing and negative imagery to reason with Pt. A, who was
making a case not to lower her opioid dosage. Figure 3 offers a visual summary of the lists found
in (I).
(I)
I1 Pt. A:

I2 Dr. A:

I3 Pt. A:
I4 Dr. A:

I5 Pt. A:
I6 Dr. A:
I7 Pt. A:
I8 Dr. A:

I9 Pt. A:

Um, I just want something for I can get up and be on my legs. And that’s all I'm
asking; I’m not asking for a lot. But, you know what? I might be scared that you
might discontinue them [Norco], but I want to be able to walk. That’s the only thing.
Right. So, I have a few options, okay? So, number one, ... we really want to know
why you’re having pain... If we don’t fix the root cause, then, you know, you’re
going to be on pain meds for the rest of your life. That’s one thing.
Mhm.
The second thing is ... opiates can make you tired, groggy, it could suppress your
breathing,... It could be a big problem. And even being on a certain amount of
opiates, there’s a risk of sudden death as well, which have been shown in multiple
studies. So one thing to do is make sure we control your pain with the right
medications, but not have you be sedated, groggy, and also have that risk for sudden
death, right? I mean that’s a scary thing.
Mhm.
So ... buprenorphine. It doesn’t suppress your respiratory drive, your breathing, it
has a good effect on pain control ... lower risk of sudden death
Mhm.
Is that something that you might be interested in converting to? Buprenorphine?
...good for you in terms of pain ...less of the side effects of the opiates and- reduce
your risk sudden death as well.
Um, yeah, we can do that… next month.

23

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

Figure 3: Illustration of lists found in Excerpt (I)

This complex exchange starts with Pt. A belaboring the importance of continuing her
opioid medication with the help of quantifying terms—“I only want one thing”, “that’s all I’m
asking”, “not asking for a lot”, and “that’s the only thing”. Pt. A also repeatedly used the verb
want in a single turn to describe (1) the opioids that help her get out of bed and (2) the ability to
walk painlessly. Combining want with words that denote ability, like can and able, suggests that
the patient is associating opioids with her capacity to walk, leaving the physician the challenging
task of reducing opioid prescriptions without seeming as if they are depriving the patient the
ability to walk.
In I2, Dr. A mirrored Pt. A’s use of quantifying terms and organized their arguments in a
list: “I have a few options, “number one”, “that’s one thing”, “the second thing”. The perception
of thoroughness associated with organized lists could also add some weight to the response. In
this case, Dr. A’s response suggests that they find giving multiple reasons the appropriate
response for convincing Pt. A. Otherwise, Dr. A would have stopped at the first reason they gave
in I2. As Jefferson (1990) pointed out, lists allow speakers to evoke a sense of credibility.
24

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

Embedded within the primary list are sublists offering additional support to the
arguments being made. For instance, the physician’s enumeration of the six opioid side effects in
I4 provides more context to their second main point, conveying the sense of thoroughness
unlikely to come with the short and singular alternative. Another observation of sublists is their
similarity to written lists in terms of pauses (commas) and coordinating conjunctions (or and
and). This may be due to sublists being comprised of entities with apparent connections that only
make sense if they appear together, within a turn, as opposed to the overarching list that goes on
across turns.
The physician’s word choices could also contribute or trigger certain images and
emotions that could motivate patients to wean off opioids or to be open to finding alternative
treatment. For example, aside from medical jargon inaccessible to patients, Dr. A listed the side
effects with phrases like big problem, right medication, scary thing, shown in multiple studies,
and make sure you’re safe. It is also noteworthy that the physician always followed the word
sudden with death. Since sudden events are likely unplanned, Dr. A is suggesting that a fatal
overdose could occur without necessary warnings, even when the patient feels safe or secure
about their dose.
4.6. Tag questions
Interrogatives like you know?, right?, okay? and remember?, which are found in many of
the excerpts presented thus far, are called tag questions. According to Lakoff (1973), tag
questions elicit approval and collaborative thinking at times of tentativeness due to the speaker’s
lack of confidence and certainty. In this study, physicians’ tentativeness does not come from lack
of policy knowledge but from the uncertainty of how patients may react to said policies. As for

25

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

collaborative thinking, there were examples in which physicians used you know? to engage
patients in collaborative guesswork, eliciting patients to fill in the blanks and finish their turns.
Take, for example, the following examples in which the physicians bring up the benefits of
naloxone regarding heroin overdose.
Dr. A: if there’s concern for overdosing... like, you know?, other substances like heroin or
anything like that, other things that may, you know? It’s still important to have this just
in case.
Dr. B: It’s [naloxone] a medication, a nasal spray, that reverses an opioid overdose.. It also
works for heroin overdoses, but, you know?

Here, both physicians used tag questions when bringing up heroin. Particularly after the
words like, may, and but, which meant the utterance in progress had not been completed. The
physicians are using tag questions in these examples to tap into some shared knowledge that they
know is recognizable to the patient. If patients confirm through words or gestures that they
understand what "you know" refers to, then the physicians may not need to say what is too facethreatening to mention or at least mitigate the tension if they decide to finish their statement.
4.7. Downtoners
Similar to tag questions, downtoners (i.e., words that mitigate the semantic weight of
face-threatening discussions like just, maybe, some, and kind of) were also a recurring
observation in the excerpts presented in this study. Here again are some of them:
Dr. A:
Dr. B:
Dr. B:
Dr. B:
Dr. C:

Can you get your lab checked real quick upstairs?
maybe you have time to go upstairs to do some quick labs?
It [drug test] should be really quick. It’s just, you know, it’s just, it’s our policy
It’s- It’s just rules, you know? ... It’s just routine, unfortunately.
We just need to do that [drug test] every so often.

26

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

In these examples, the physicians used just to inject a level of mundanity into their
mandates and downplay the significance of having to undergo routine tests, receive naloxone,
and discuss alternative treatments. Quirk et al. (1985) describe downtoners as a specific form of
intensifier aimed at decreasing a proposition’s intensity and, therefore, minimizing the
significance of a particular FTA (see also Aijmer, 2002; Lee, 1987). In line with this definition,
this study also considers phrases intended to modify time (e.g., every so often, quick labs, real
quick, and sometime, among others) as downtoners minimizing the role of time to solicit
cooperation. In this study, references of time occurred during discussions of lab testing. By
specifying that the lab tests are regular and quick, the physicians are able to impress upon their
patients that drug tests are common and untroubling.

5. Discussion
This study investigates the face-saving features used by physicians while enacting
restrictive opioid policies. The question is motivated by the physician’s dilemma that simply
enacting policies can be perceived by patients as distrusting, shaming, and suspecting of their
pain (Dassieu et al., 2021). Through DA, the observed communicative practices unveiled
physicians’ constant negotiation between maintaining collaborative relationships and policy
demands. A total of eight face-saving features were found in the study. Table 4 provides a
summary of the face-saving features physicians employed and the potential roles such discourse
features play in opioid discourse.

27

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

Table 4: Summary of face-saving acts
Face-saving acts

Description

Potential role in opioid discourse

Pseudo requests
(Benkendorf et al.,
2001)

Reframing imperatives and
mandates into suggestions,
questions, or requests.

To frame treatment decisions as
collaborative.

Downtoners
(Quirk et al., 1985)

Using lexical items that decrease
the intensity of the message, such
as just, maybe, or kind of.

To mitigate potential consequences and
avoid causing panic and distress.

Impersonalization
(Rundblad, 2007)

Using an ambiguous reference for
actions or events supposedly for
the listener, as if they were talking
about someone else.

To discuss overdose and dependence
without referencing the patient.

Broadening Audience
(Brown & Levinson,
1987)

Explaining that listeners are
subject to certain actions as a
result of their group membership
and not a malicious, targeted, or
isolated event.

To signal authority and credibility for
decisions and suggestions that affect the
patient.

Broadening - Scope
(Brown & Levinson,
1987)

Distancing one’s participation or
sole contribution to an event or a
decision, such as the use of the
inclusive words we and our.

To signal collaborative decision-making
or past agreement toward a proposition.

Broadening Interpretation (Torres,
2021)

Using modal verbs such as may or
would to broaden the interpretive
spaces of propositions.

To introduce some sense of optionality
mitigating the intensity of a command.

Tag questions
(Lakoff, 1973)

Asking confirmation questions,
such as Okay? and You know? to
seek approval and confirmation.

To signal collaborative thinking, bring
up supposedly recognizable
information, and maintain alignment.

Listing
(Dori-Hacohen, 2020;
Jefferson, 1990)

Organizing information to convey
thorough knowledge of the topic.

To enumerate credible justifications
behind policy demands.

Redirection
(Friedlander et al.,
2000)

Diverting conversations toward
alternative paths such as
discussing the source of pain
instead of pain itself.

To frame pain as an entity that
originates from a particular source and
is therefore treatable.

Referencing policies to minimize
one’s involvement with policy
demands

To and rules to maintain collaborative
relationships in moments where patients
disagree with the suggested treatment.

Shaping the ways in which hearers
imagine and make meaning out of
images.

To further justify policy enactments and
add convincing power to treatment
suggestions.

(Negative) Imagery
(Tannen, 2007)

Note: Collectively, these features share one common purpose: maintaining collaborative relationships and preempting arguments
or disagreements.

28

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

Aside from the primary results, this study also expands on existing literature focusing on
discourse features, face-work, and doctor-patient interactions in the following ways:
1. Lists indicate plurality, which makes listing a suitable discourse feature for
communicating legitimacy, adding a sense of fullness, or highlighting the multitude of
what is being said. Spoken lists further benefit from the use of sequencing verbiage such
as number one and the second thing is.
2. In line with Quirk et al.’s (1985) definition of downtoners as expressions that downplay
FTAs, phrases such as every so often and real quick are downtoners by function because
they minimize the role of time needed for drug tests.
3. Highly technical and unexplained scientific language or medical jargon can contribute to
(negative) imagery.
4. Modals are just as much a broadening tool in spoken interactions as previous studies have
observed in written discourse (Torres, 2021).
5. Impersonalization, a process that has been previously documented in written health
discourses (Rundbland, 2007), is just as salient in verbal medical exchanges.
6. Impersonalization, a process commonly associated with passive construction in written
text, emerges in myriad other ways—from repairs to wordplay—in spoken interactions.

As can be gleaned from the findings, the general analytic approach used here is in line
with contemporary approaches that consider the relational or interpersonal nature of face (i.e.,
Locher and Watts, 2005). The context in which these interactions occurred helped pinpoint the
potential FTAs and the strategies used to address them. For example, patient responses are one of
many factors that could affect physician strategies; as shown in the physicians' use of audience

29

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

generalization and redirection whenever patients push back in Excerpts C and H, respectively. In
other words, context aids in elucidating probable reasons for physicians' linguistic choices.
Spencer-Oatey (2008, p. 37) also names "sociality rights and obligation"—the right to expect
certain outcomes and the obligation to carry out certain actions—as an essential contextual factor
for maintaining harmonious relationships. This research shows how cooperative relationships
may be jeopardized when patients' expectations and doctors' obligations conflict. When it comes
to discussions about opioids, patients assert their right to treatment, while physicians confront the
conundrum of denying these rights to fulfill their obligation not to endanger their patients.
Since face-work is about mitigating threats and seeking agreement, it seems necessary to
discuss patient compliance. Patients were non-compliant with trying alternative treatments and
agreeing to reduce opioid dosage. When it came to the drug screening, Pt. A did not comply,
while Pt. B agreed after initially pushing back. In perhaps the smoothest transaction in this study,
Pt. C complied with the screening without any contentions. As can be gleaned from the excerpts,
the discussion of naloxone was met with pushback from all patients, but since the physicians
prescribed them, this study does not have the information on whether they picked up the
medication. On that note, further investigation using a more extensive data pool could help
uncover which features, or permutations thereof, effectively meet policy demands or result in
compliance. Doing so could offer solutions to common quandaries such as choosing whether to
give orders through pseudo requests, which was prone to rejection in this study, or through
unmitigated imperatives. The physicians noted that they often get briefed on new policies but do
not consistently receive guidance on communicating them. Another factor potentially at play
here is the presumption that comes with handing physicians, whose reputations include enduring
years of medical school, a printout of sample dialogues. Further investigations are needed to

30

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

examine physicians’ sensitivity when receiving guidance and whether it would be more effective
and efficient if the current memos they receive are replaced with equally informative, ready-touse templates that physicians can personalize.
Although the taxing nature of enacting opioid policies can be easily gleaned from the
presence of false starts, paraphrasing, and repairs in the transcript, future investigations should be
undertaken to describe the prosodic and suprasegmental features associated with face-work in
discourses concerning health and policy issues. While linguistic backgrounds were controlled for
in this study, analyzing a larger corpus of interactions with additional emphasis on cultural
identities and structures is important, as it can offer an alternative lens through which healthcare
policies and interactions involving controlled substances could be examined. The findings of this
study are encouraging and should be validated by a larger sample size in future investigations.
The prospect of improving how institutions could assist policy stakeholders (e.g., medical
professionals and front-line workers) in the discursive aspects of enacting policies serves as a
continuous incentive for future research in this space.
6. Conclusion
By identifying physicians’ face-saving strategies, this study calls attention to the
challenging nature of having to implement opioid policies without jeopardizing collaborative
partnerships with patients. Looking at the bigger picture, the socialization process physicians
must endure every time a new policy emerges could be streamlined, either at the state policy
level (through the codification of directives requiring institutions to offer communication
support) or at the institutional and departmental levels (by providing some guidance on
communicating policies to patients). Most policies focus only on the What? while brushing off
the How? of policy implementation. Policymakers have demanded that physicians reduce their
31

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

opioid prescribing but have not acknowledged the difficulty of figuring out the best way to
broach this topic with patients. How do physicians enact policies if their compliance leads to
patients' non-compliance? This study corroborates the arguments made by policy researchers—
including Ramanathan (2005) and Canagarajah (2005)—that it is through the examination of
policies from a bottom-up perspective that the disconnect between policy demands and local
enactments emerges.
The need for improved communication opens opportunities for collaborations between
medical institutions that adapt state policies into their practices, physicians who interact with
chronic pain patients, and linguists whose limited exposure to the space provide a fresh yet
critical perspective. Such a practical approach to enacting policies could indeed improve the
efficacy of pain management and minimize the amount of face-work performed by physicians.

32

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

References:

Aijmer, K., 2002. English Discourse Particles: Evidence from a Corpus, Studies in corpus
linguistics. John Benjamins Publishing, Amsterdam.
Ainsworth-Vaughn, N., 1998. Claiming power in doctor-patient talk. Oxford University Press,
New York.
Al Achkar, M., Revere, D., Dennis, B., MacKie, P., Gupta, S., Grannis, S., 2017. Exploring
perceptions and experiences of patients who have chronic pain as state prescription
opioid policies change: a qualitative study in Indiana. BMJ Open 7, e015083.
https://doi.org/10.1136/bmjopen-2016-015083
Arborelius, E., Thakker, K.D., 1995. Why is it so difficult for general practitioners to discuss
alcohol with patients? Fam Pract 12, 419–422. https://doi.org/10.1093/fampra/12.4.419
Aronsson, K., Sätterlund-Larsson, U., 1987. Politeness Strategies and Doctor-Patient
Communication. On the Social Choreography of Collaborative Thinking. Journal of
Language and Social Psychology 6, 1–27. https://doi.org/10.1177/0261927X8700600101
Bogdanowska-Jakubowska, E., 2010. Face: An interdisciplinary perspective. Wydawnictwo
Uniwersytetu Śląskiego.
Beach, W. A., 1993. Transitional regularities for ‘casual’ “Okay” usages. Journal of Pragmatics
19, 325–352.
Benkendorf, J.L., Prince, M.B., Rose, M.A., Fina, A.D., Hamilton, H.E., 2001. Does indirect
speech promote nondirective genetic counseling? Results of a sociolinguistic
investigation. American Journal of Medical Genetics 106, 199–207.
https://doi.org/10.1002/ajmg.10012
Brown, P., Levinson, S.C., 1987. Politeness: some universals in language usage, 21. pr. ed,
Studies in interactional sociolinguistics. Univ. Press, Cambridge.
Brown Jr, E. G., Sewell, D. S., Kirchmeyer, K., 2014. Guidelines for prescribing controlled
substances for pain. https://www.ombc.ca.gov/forms_pubs/pain_guide.pdf
Caffi, C., 1999. On mitigation. Journal of Pragmatics, Pragmatics: The Loaded Discipline? 31,
881–909. https://doi.org/10.1016/S0378-2166(98)00098-8
Canagarajah, A.S. (Ed.), 2005. Reclaiming the local in language policy and practice, ESL and
applied linguistics professional series. L. Erlbaum Associates, Mahwah, N.J.
Clark, S.J., Hudak, P.L., 2011. When Surgeons Advise Against Surgery. Research on Language
& Social Interaction 44, 385–412. https://doi.org/10.1080/08351813.2011.619313
33

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

Culpeper, J., 2011. Politeness and impoliteness, in: Andersen, G., Aijmer, K. (Eds.), Pragmatics
of Society, Handbooks of Pragmatics. Walter de Gruyter, Berlin, pp. 393–438.
Dassieu, L., Heino, A., Develay, É., Kaboré, J.-L., Pagé, M.G., Hudspith, M., Moor, G.,
Choinière, M., 2021. Conversations About Opioids: Impact of the Opioid Overdose
Epidemic on Social Interactions for People Who Live With Chronic Pain. Qual Health
Res 10497323211003064. https://doi.org/10.1177/10497323211003063
Dori-Hacohen, G., 2020. The “Long List” in oral interactions: Definition, examples, context, and
some of its achievements. PRAG 30, 303–325. https://doi.org/10.1075/prag.19007.dor
Epstein, R., Morse, D., Frankel, R., Frarey, L., Anderson, K., Beckman, H., 1998. Awkward
Moments in Patient-Physician Communication about HIV Risk. Annals of internal
medicine 128, 435–42. https://doi.org/10.7326/0003-4819-128-6-199803150-00003
Evans-Agnew, R.A., Johnson, S., Liu, F., Boutain, D.M., 2016. Applying Critical Discourse
Analysis in Health Policy Research: Case Studies in Regional, Organizational, and
Global Health. Policy, Politics, & Nursing Practice 17, 136–146.
Friedlander, M.L., Heatherington, L., Marrs, A.L., 2000. Responding to blame in family therapy:
A constructionist/narrative perspective. The American Journal of Family Therapy 28,
133–146.
Gee, J.P., 2011. An introduction to discourse analysis: theory and method, 3. ed. ed. Routledge,
Milton Park, Abingdon.
Goffman, E., 1955. On face-work: An analysis of ritual elements in social interaction. Psychiatry
18, 213–231.
Henry, S.G., Bell, R.A., Fenton, J.J., Kravitz, R.L., 2017. Goals of Chronic Pain Management.
The Clinical journal of pain 33, 955–961.
Henry, S.G., Chen, M., Matthias, M.S., Bell, R.A., Kravitz, R.L., 2016. Development of the
Chronic Pain Coding System (CPCS) for characterizing patient-clinician discussions
about chronic pain and opioids. Pain Medicine 17, 1892–1905.
Henry, S.G., Gosdin, M.M., White, A.E., Kravitz, R.L., 2019. “It Sometimes Doesn’t Even
Work”: Patient Opioid Assessments as Clues to Therapeutic Flexibility in Primary Care.
Journal of general internal medicine 35, 1635–1640.
Holman, H., Lorig, K., 2000. Patients as partners in managing chronic disease: partnership is a
prerequisite for effective and efficient health care. BMJ 320, 526–527.
Hood-Medland, E.A., White, A.E., Kravitz, R.L., Henry, S.G., 2021. Agenda setting and visit
openings in primary care visits involving patients taking opioids for chronic pain. BMC
Family Practice 22, 1–11. https://doi.org/10.1186/s12875-020-01317-4
34

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

Hughes, H.K., Korthuis, P.T., Saha, S., Eggly, S., Sharp, V., Cohn, J., Moore, R., Beach, M.C.,
2015. A mixed methods study of patient–provider communication about opioid
analgesics. Patient Education and Counseling 98, 453–461.
https://doi.org/10.1016/j.pec.2014.12.003
Jefferson, G., 1990. List construction as a task and resource. Interaction competence 63, 92.
Johnstone, B., 2018. Discourse analysis. John Wiley & Sons.
Kampf, Z. 2015. Political discourse analysis. International encyclopedia of language and social
interaction 1–17.
Karafoti, E., 2021. Negotiating preferred norms in requests and offers: Is the (dis)preferred
answer so obviously (im)polite? Journal of Pragmatics 173, 134–147.
https://doi.org/10.1016/j.pragma.2020.07.010
Klamen, D.L., 1999. Education and Training in Addictive Diseases. Psychiatric Clinics of North
America 22, 471–480. https://doi.org/10.1016/S0193-953X(05)70087-3
Lakoff, R., 1973. Language and woman’s place. Language in society 2, 45–79.
Lee, D., 1987. The semantics of just. Journal of Pragmatics 11, 377–398.
https://doi.org/10.1016/0378-2166(87)90138-X
Locher, M.A., Watts, R.J., 2005. Politeness Theory and Relational Work. Journal of Politeness
Research. Language, Behaviour, Culture 1. https://doi.org/10.1515/jplr.2005.1.1.9
Lutfey, K., Maynard, D.W., 1998. Bad News in Oncology: How Physician and Patient Talk
About Death and Dying Without Using Those Words. Social Psychology Quarterly 61,
321–341. https://doi.org/10.2307/2787033
Martínez, G., 2008. Language-in-healthcare policy, interaction patterns, and unequal care on the
US-Mexico border. Language Policy 7, 345–363.
Matsumoto, Y., 1988. Reexamination of the universality of face: Politeness phenomena in
Japanese. Journal of pragmatics 12, 403–426.
Matthias, M.S., Krebs, E.E., Collins, L.A., Bergman, A.A., Coffing, J.M., Bair, M.J., 2013. “I’m
Not Abusing or Anything”: Patient–physician communication about opioid treatment in
chronic pain. Patient education and counseling 93, 197–202.
Mistral, R., Velleman, W., 2001. Substance-misusing Patients in Primary Care: incidence,
services provided and problems. A survey of general practitioners in Wiltshire. Drugs:
Education, Prevention and Policy 8, 61–72. https://doi.org/10.1080/09687630123912

35

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

O’Driscoll, J., 2017. Face and (Im) politeness, in: Culpeper, J., Haugh, M., Kádár, D.Z. (Eds.),
The Palgrave Handbook of Linguistic (Im) Politeness. Palgrave Macmillan UK, London,
pp. 89–118. https://doi.org/10.1057/978-1-137-37508-7
Quirk, R., Greenbaum, S., Leech, G., Svartoik, J., 1985. A Comprehensive grammar of the
English language. Longman, London ; New York.
Ramanathan, V., 2005. Rethinking language planning and policy from the ground up:
Refashioning institutional realities and human lives. Current Issues in Language Planning
6, 89–101.
Ramanathan, V., 2010. Introduction to thematic issue: language policies and health. Language
Policy 9, 1–7. https://doi.org/10.1007/s10993-009-9154-7
Roberts, C., Sarangi, S., 2005. Theme-oriented discourse analysis of medical encounters.
Medical education 39, 632–640.
Roberts, F., Kramer, J.S., 2014. Medication and morality: Analysis of medical visits to address
chronic pain, in: Hamilton, H., Chou, W.S. (Eds.), The Routledge Handbook of Language
and Health Communication. Rouletge, London, pp. 477–489.
Rundblad, G., 2007. Impersonal, General, and Social: The Use of Metonymy Versus Passive
Voice in Medical Discourse. Written Communication 24, 250–277.
https://doi.org/10.1177/0741088307302946
Sacks, H., Schegloff, E.A., Jefferson, G., 1978. A simplest systematics for the organization of
turn taking for conversation, in: Studies in the Organization of Conversational
Interaction. Elsevier, pp. 7–55.
Schilling-Estes, N., 1998. Investigating “self-conscious” speech: The performance register in
Ocracoke English. Language in society 27, 53–83.
Sifianou, M., Tzanne, A., 2021. Face, Facework and Face-Threatening Acts, in: Kádár, D.Z.,
Terkourafi, M., Haugh, M. (Eds.), The Cambridge Handbook of Sociopragmatics,
Cambridge Handbooks in Language and Linguistics. Cambridge University Press,
Cambridge, pp. 249–271. https://doi.org/10.1017/9781108954105.015
Spencer-Oatey, H., 2008. Face, (im)politeness and rapport, in: Spencer-Oatey, H. (Ed.),
Culturally Speaking: Culture, Communication and Politeness Theory. Continuum
International Publishing Group, London, U.K. / New York, NY, U.S.A., pp. 11–47.
Tannen, D., 2007. Talking Voices: Repetition, Dialogue, and Imagery in Conversational
Discourse. Cambridge University Press.

36

Accepted Author Manuscript Accepted Author Manuscript Accepted Author Manuscript

Terkourafi, M., 2012. Politeness and pragmatics, in: Allan, K., Jaszczolt, K.M. (Eds.), The
Cambridge Handbook of Pragmatics. Cambridge University Press, Cambridge, pp. 617–
637. https://doi.org/10.1017/CBO9781139022453.034
Ting-Toomey, S., Kurogi, A., 1998. Facework competence in intercultural conflict: an updated
face-negotiation theory. International Journal of Intercultural Relations 22, 187–225.
https://doi.org/10.1016/S0147-1767(98)00004-2
Torres, P.J., Henry, S.G., Ramanathan, V., 2020. Let’s talk about pain and opioids: Low pitch
and creak in medical consultations. Discourse Studies 22, 174–204.
https://doi.org/10.1177/1461445619893796
Torres, P.J., 2021. "The Role of Modals in Policies: The US Opioid Crisis as a Case Study”.
Applied Corpus Linguistics. https://doi.org/10.1016/j.acorp.2021.100008
Torres, P.J., 2022a. “Modality and interpretative spaces in policies”. In B.A. Diaz & R.W.
Schrauf (Eds). Applying linguistics in health research, education, and policy: Bench to
bedside and back again. De Gruyter Mouton. https://doi.org/10.1515/9783110744804006
Torres, P.J., 2022b. “The Nation’s Fix: The Language of the Opioid Crisis”. The Nation’s Fix:
The Language of the Opioid Crisis. (University of California, Davis, 2022).

van Boekel, L.C., Brouwers, E.P.M., van Weeghel, J., Garretsen, H.F.L., 2013. Stigma among
health professionals towards patients with substance use disorders and its consequences
for healthcare delivery: Systematic review. Drug and Alcohol Dependence 131, 23–35.
https://doi.org/10.1016/j.drugalcdep.2013.02.018
Van Dijk, T.A., 1999. Critical discourse analysis and conversation analysis. Discourse & Society
10, 459–460.
Watts, R.J., 2003. Politeness. Cambridge University Press.
Wilkinson, S., Kitzinger, C., 2000. Thinking differently about thinking positive: a discursive
approach to cancer patients’ talk. Social science & medicine 50, 797–811.
Woods, N., 2014. Describing Discourse: A Practical Guide to Discourse Analysis. Routledge.

37