“Having an Ovary this Big is Not Normal”: Physicians’ Use of Normal to Assess Wellness and Sickness During Oncology Interviews

Abstract

During oncology interviews physicians and patients routinely employ the term normal to describe patients’ condition and overall health status. Surprisingly prevalent but little understood, normal is recruited to achieve an array of social actions comprising a primal aspect of patient-provider interactions: Determining, assessing, and treating patients’ health status as well and/or sick. Utilizing conversation analysis (CA) to examine a collection of 136 normal references across 61 oncology interviews, this paper draws from a sub-sample of 101 instances to examine how physicians use normal to perform four specific sets of social actions: 1) Invoking normal as a preferred range, 2) Utilizing normal as evidence that does not explicitly label patients’ conditions, 3) Treating the absence of normal as indicative of sickness, and 4) Providing reassurance to patients in the presence of normal and not normal circumstances. Special attention is given to how oncologists make the case for wellness and/or sickness by identifying patients as normal or not normal/abnormal. Future research is needed to understand how patients employ normal to characterize their symptoms and negotiate health status, explore how or if alternative orientations to normal are based on gender and/or varying ethnicities, and to develop a more precise understanding of alternative practices for describing and categorizing test results (e.g., common/uncommon).

Determining what counts as normal is fundamental to understanding, diagnosing, and treating cancer. Physicians employ the term “normal” to situate patients’ diagnosed conditions within baseline values of wellness and sickness. During clinical encounters, we examine moments where physicians invoke normal in the midst of accomplishing important social actions: Assessing test results, recommending courses of treatment, and explaining for patients how normal or abnormal evidence indicates the presence, absence, and potential threat of cancer. What follows are 1) a brief discussion of prior research focusing on normality, wellness, and sickness in medicine and cancer care, 2) a description of the data and methods for this study, 3) analysis of primary social actions comprising physicians’ references to normal, and 4) a summary of findings and implications for future research.

Situating Normal in Relation to Wellness and Sickness

Historically, different and often contradictory conceptions of what it means to be normal (e.g., vs. deviant or pathological, see Durkheim, 1982) are scattered throughout a diverse body of literature. Fundamentally, normal functions as an expectation that others and society as a whole will act ordinarily (Swaan, 1999). The overwhelming tendency to cling to normalcy when managing everyday affairs, including during crises, was emphasized as Sacks (1984) observed “it is almost everybody’s business to be occupationally ordinary” (p. 419). In one of the only studies to specifically examine the use of the term normal, Bredmar and Linell (1999) identified how references to normal were typically positive, possess both moral and statistical dimensions, and at times get designed to offer reassurance. Normal is thus often framed as a good and preferred state-of-affairs. Within organizational cultures, for example, managers employ personality tests and generate other criteria to discern normal from abnormal individuals for purposes of hiring, firing, and promotion (Nadeson, 1997).

Normal and Medical Care

Alternative conceptions of normal have figured prominently in the history and study of medical care. Over 50 years ago, Wellman (1958) urged medical practitioners to recognize that conceptions of normal, while essential for medical care, are inherently complex and hold the potential to create communication difficulties. Multiple meanings get attributed to such a common word, and physicians routinely employ normal in different and often contrasting ways: To characterize the lack of medical problems, the most usual circumstances (i.e., as standard and not deviant), the middle of a statistical distribution, or even a perfect and ideal health condition. What counts as normal also varies according to how standardized medical practices define any given behavioral domain within a given disease, family, sub-culture, or society.

Depictions of the sick role by Parsons (1951, 1978) draw attention to how it is normal for sick patients to have certain rights and obligations (e.g., a legitimate excuse for missing work or school). Yet, it is also normal for the sick role to be exploited (e.g., by falsely reporting sickness). This type of exploitation qualifies as sanctioned deviance that can challenge and burden a productive society. Yet normal patients are often considered to be those who are not diagnosed as sick (Rose, 2011). In contrast, when patients experience varying degrees and types of sickness, each malady (e.g., common cold) has its own range of normal symptoms (e.g., chills, runny nose, and a cough) and normal treatment protocol (e.g., rest and plenty of fluids).

As Engel (1977) noted, it is the “historic function of the physician to establish whether the person soliciting help is ‘sick’ or ‘well’” (p.18) and to best treat the sickness. When physicians and patients work together to determine the patient’s identity as well and/or sick (Cloute, Mitchell, & Yates, 2008; Maynard & Heritage, 2005; Rapley, Kiernan, & Anataki, 1998), they warrant their claims through resources such as the evidence of tests (e.g, see Peräkyla, 1998, 2002), symptoms and observable problems, medical history, population ranges, and even anecdotal experiences. Like a dimmer-switch, however, what counts as normal can vary considerably across these alternative modes of information. Speakers do not always invoke normal to represent a binary category. Instead, as with okay (see Beach, 1993; Pillet-Shore, 2003), normal may involve a gradated metric spanning from normal to not normal, and from wellness to sickness. Focusing on parent-teacher conferences, Pillet-Shore (2003) identified how an okay assessment can be understood as binary (e.g., good or bad) or gradated to indicate varying dimensions of students’ performances. Physicians and patients may also use normal as a gradated metric, with greater or lesser degrees of normality associated with wellness and/or sickness.

Understanding Normal Practices in the Midst of Cancer Care

As Wellman (1958) noted, normal is a moving target that is not easily assigned a stable definition or usage. Throughout cancer care, what is normal for a healthy person (e.g., control of bowel movements, absence of acute pain) may nevertheless result in cancer diagnosis given the presence of other symptoms and test findings. Similarly, what may be normal for a diseased person (e.g., lack of bowel control, presence of acute pain) does not automatically translate into evidence for cancer. As interaction unfolds, usages of normal can thus be fraught with ambiguities.

One set of tasks faced by oncologists involves educating patients about matters such as how normal cells become cancerous when particular genes alter cellular growth and development, as well as the progressive reduction or growth of tumors. Cancerous tumors are formed when uncontrolled cells form into masses (with the exception of leukemia). It is normal for cancer to involve uncontrolled division of cells that, left untreated, can lead to invasive organ damage and spreading to other body parts (i.e., through metastasis via bloodstream or the lymphatic systems). Biopsy and imaging results reveal benign or malignant cells, the latter being more aggressive and indicating sickness that, without treatment, may progressively threaten quality and length of life – which is normal for many types of cancer.

Within cancer clinics, physicians and patients have been shown to use varying forms of evidence to claim certain qualities or identities for themselves and others (Lindlof & Taylor, 2002; Beach, 2013a). Oncology interviews can be understood as a negotiation process between patients and physicians as different stances are taken, and varying medical evidence is discussed, to determine what counts as normal in the patient’s body (Rapley et al., 1998). But how do these moments get enacted and organized into encompassing patterns of communication activities?

Consider an initial example, drawn from an interview between a new patient who has been referred to a surgical oncologist within a comprehensive cancer center. This exchange involves the physician and the patient’s daughter who accompanied her parent to the clinic:

Excerpt 1 (ND14P1:13-14)¹

1	Physician:	→ The fact that the other ovary looks normal, the fact that we don’t see any other
2		areas that look like they might be cancer. Nonetheless, in a woman of seventy
3		→ years, having an ovary this big is not normal.=
4	Daughter:	Okay.
5	Physician:	=And that’s why I think Dr. G. has recommended surgery, =
6	Daughter:	Okay.
7	Physician:	=and that’s why I probably, you know, I’ll end up recommending surgery too
8		after I do a physical exam.

The first reference (line 1) describes one of the patient’s ovaries as “normal,” employed here as evidence that cancer may not exist. The physician’s second reference (line 3), however, treats a “big” ovary as “not normal”. The physician uses this assessment as a rationale for a likely recommendation for surgery (lines 7–8). For these instances, normal/not normal were employed to distinguish between two ovaries: One apparently normal and thus healthy, and the other (“big”) revealing suspicious signs of cancer that surgery may (or may not) confirm. In Excerpt 1 (above), then, the physician comments that the patient is elderly, has an ovary that is larger than normal, and uses this evidence to imply that the patient is sick, requires treatment, and thus is a candidate for surgery.

As Excerpt 1 (above) only begins to reveal, being normal is not simply something cancer patients “are” (Widdicombe, 1998b), but a set of identifiable practices for constructing social activities through talk and embodied social actions (e.g., Antaki, Condor, & Levine, 1996; Greatbatch & Dingwall, 1998; Heritage & Clayman, 2010; McKinley & Dunnett, 1998; Zimmermann, 1998). To better understand variable usages and consequences of normal involves discovering the particular interactional and sequential environments when references to normal occur, the primary social actions comprising these moments, and how the organization of these actions are consequential for understanding how wellness and sickness get co-constructed. The diverse and gradated functions of normal throughout diagnosis and treatment, however, have not been systematically investigated and a need exists to reveal how wellness and sickness get interactionally formulated (Greatbatch & Dingwall, 1998; Goetz, 2008). Within several studies of physician-patient interaction, normal appears in the data but has generally remained unaddressed in favor of other important interactional features (See: Frankel, 1984, Excerpt 2; Roca-Cuberes, 2008, p. 555). Thus, crucial aspects of managing wellness and sickness remain unexamined.

In cancer clinics, close examinations of patient-provider interactions can yield important, even counter-intuitive findings. For example, it has been shown that cancer patients invest considerable efforts justifying their wellness to oncologists (Beach, 2013a, b) – actions which may first appear to be counter-intuitive, until it is recognized that patients’ pursuits of wellness are important resources for countering threatening concerns and, by so doing, managing hopeful approaches useful for overcoming sickness. A need thus exists to closely examine interactions in cancer clinics, and to lay bare numerous usages and ambiguities surrounding normal, wellness, and sickness that would not otherwise be understood as essential for seeking and providing medical care.

The ensuing analysis examines normal usages in oncology interviews, and explores how physicians invoke an array of social actions during routine visitations with new and returning cancer patients. It will be shown that when normal gets explicitly invoked, physicians assess the clinical status of normal/not normal, communicate their orientations to patients as the evidentiary basis for their diagnosis (e.g., see Maynard, 2004; Peräkyla, 1998, 2002). By doing so, as with Excerpt 1 (above), physicians retain authority to pursue medical solutions to problems.

Four specific social actions performed by physicians during oncology interviews are examined below: 1) Invoking normal as a preferred range, 2) Utilizing normal as evidence that does not explicitly label patients’ conditions, 3) Treating the absence of normal as indicative of sickness, and 4) Providing reassurance to patients in the presence of normal and not normal circumstances.

Data and Method

Oncology encounters were video recorded at an NIH/NCI designated comprehensive cancer center. Small digital cameras, mounted on tripods, recorded interviews without the presence of researchers. Full IRB approvals from participating institutions were obtained, including completion of voluntary consent forms by patients and oncologists. Participants’ names have been made anonymous and confidential.

Conversation analysis (CA) is employed to analyze a collection of 136 instances, across 61 oncology interviews, when normal gets explicitly invoked by patients and physicians. Data for this paper are drawn from a sub-sample of 101 instances in which physicians invoke normal. Video recordings were carefully transcribed, and both transcriptions and recordings were repeatedly examined to reveal how moments involving normal references were organized (see Drew & Heritage, 2006; Maynard & Heritage, 2005). Emphasis is given to how participants work together to produce and manage social actions shaping unfolding interactions (Atkinson & Heritage, 1984; Heritage & Maynard, 2006; Sidnell & Stivers, 2013). CA is thus a rigorous method for providing warrantable claims that social actions are “grounded in the conduct of the parties, not in the beliefs of the writer” (Schegloff, 2007, p. 476). Functioning as a bridge between often disparate methodologies of ethnography and quantitative research (Maynard & Heritage, 2005), CA explicates how patients and physicians rely on talk-in-interaction and embodied behaviors (e.g., gaze, gesture, body orientations, and touch) to demonstrate understandings about a wide range of matters including wellness, sickness, overall health conditions, treatment regimens, and healing possibilities (see Beach, 2013c).

Basic Interactional Features of Physicians’ Normal Usages

Invoking Normal as a Preferred Range

When physicians report test results as normal, they discuss the tests in reference to a preferred range. In the following excerpts, when physicians invoke normal they a) establish the typical range and characteristics of a potential problem (e.g., enlarged nodes, platelet counts), and b) subsequently frame patients’ condition as an attribute within that range. Physicians can then locate patients’ test results, symptoms, and medical history somewhere within the established range as a basis for assessing sickness and/or wellness.

In Excerpt 2, the physician calibrates the size of the patient’s shrinking nodes within a preferred range for healthy lymph nodes:

Excerpt 2 (OCD1P7:10)

1	Physician:	Ri::ght. Any-any place that we saw enlarged nodes before they’re smaller.
2		So that’s the first important thing. A:nd u:h, there’s the biggest node
3		that’s left is sorta deep down in here. It still measures one point six,
4		which is about like that, (0.2) and previously it had been like that ((using fingers
5		to calibrate size.)) So um but-but anything a centimeter or so (.) and less is
6	→	considered normal. So what we’ll do, we’ll just follow this particular lymph node.
7		Now d-did you have a pet scan before your transplant?

The physician first describes how enlarged nodes are now smaller, then uses a gesture of pinched fingers to portray the size of “the biggest node” (see Beach, 2012b) as a mass that “measures one point six, which is about like that” (lines 3–4), now smaller than during previous measurements. To provide a referent for these numbers, the physician next utilizes normal to establish a range in lines 5–6. Here normal is employed as a gradated metric (Pillet-Shore, 2003). Since the patient’s lymph node is 1.6 (lines 3–4), the patient’s lymph node is outside the normal range but smaller than before and approaching the normal range (less than 1 centimeter). In this way, the physician establishes the range and then situates the patient’s condition in reference to this normal range by making clear that with a 1.6 centimeter node a health threat does not exist but is something “we’ll just follow” (line 6). By establishing a normal range (at or below 1 centimeter), and by comparing the 1.6 sizing of a node, the patient is informed that close monitoring will occur in contrast to more troubling news requiring further intervention.

In the excerpt below the physician establishes a numerical range for a normal platelet count:

Excerpt 3 (OCD6P4:15)

1	Physician:	We’re going to be getting the differential here and I’ll be looking at
2		your slides or - =
3	Patient:	= What type of variation do you see with the platelets from test to test?
4	Physician:	Well you know, I think they can vary from anywhere from plus or minus
5	→	50,000 or so. It’s normal to have some over 140,000, so you have 237,000
6	Patient:	oMm hm.o
7	Physician:	And they can fluctuate by as much as 20,000 if you were to jump up
8		and down- run up the stairs or something that they could change.
9	Patient:	Really, okay.

In response to the patient’s query about the variation of platelets (line 3), the physician defines a range (lines 4–5), and describes normal as “to have some over 140,000, so you have 237,000” (line 5). At least a portion of patient’s condition far exceeds the normal range of a healthy person. Since the patient’s platelet scores are not within the normal range, based on a gradient assessment with scores more or less normal (Pillet-Shore, 2003), the patient risks exclusion from the identity of healthy person. Though the physician does not explicitly announce these scores as potentially bad news, the patient treats the scores – much higher than normal – with both surprise and resignation by responding with “Really, okay”(line 9). The patient’s “Really” seems to indicate surprise at the information while “okay” indicates a type of acceptance.

In another interview, a physician delivers news to the patient about both the patient’s individual tests and overall condition. The first item raised, the patient’s “platelet-problem” (line 1), is next tempered with the good news of an “okay” white blood count (line 2):

Excerpt 4 (OCD1P5:3)

1	Physician:	Uhh. Well I-I’ve been talking to people about your platelet- problem. (.)
2		U:m. But the white count looks ↑okay.
3	Patient:	Everything looks good except the [platelet.
4	Physician:	[Yeah. The platelet.=
5	Patient:	=Uh-huh. What is this RDW that- keeps going ↓down.
6	Physician:	That-that’s just a number that u:hh, it’s-it’s not a number that’s very useful
7		for you. It’s only going down a little bit. It’s actually suppose to be low.
8	Patient:	oOkay.o
9	Physician:	→The ↑normal number is more like 18 or 19.
10	Patient:	↑Okay. Okay.=

The patient reformulates the physician’s assessment as “everything looks good except the platelet.” (line 3), which is confirmed by the physician (line 4). In line 5, the patient’s inquiry about the RDW (red blood cell distribution width) test, which measures how much erythrocytes vary in size, indicates a lack of awareness about what these tests mean. In particular, high RDW has been tied to various health problems and increased mortality (Chen et al., 2009). The physician, relying on medical knowledge and thus in a position to assert authority about these tests (Heritage, 2009), describes how it is not an important number for the patient to be concerned with – especially because it is “suppose to be low” (lines 6–7).

The patient’s minimal and quiet response of “^oOkay.^o” (line 8) is not atypical for patients when hearing a diagnosis, or acknowledging complex information that is not well understood (Heath, 1992). The physician establishes a range of normal in line 9, with the statement “the ↑normal number is more like 18 or 19”. In this case and overall statistically, normal can be synonymous with average or normal in terms of a population bell curve. Once again, the physician is able to position the patient’s tests in relation to a normal range. This action is acknowledged by the patient with the response of “↑ Okay. Okay”(see Excerpt 4 above). Finding out that a potential source of anxiety – as with “RDW that -keeps going ↓down.” – is actually a normal range, can de-trigger the need for worry (at least about that particular test and set of results).

In summary, the previous three instances reveal how physicians rely on normal to establish a range for situating patients’ condition as not urgent but worthy of monitoring, likely requiring further intervention, or even as good news (e.g., when lower than normal scores indicate wellness rather than sickness). The physicians often define normal in reference to numbers and statistics, details that most patients will not understand without physicians’ authoritative knowledge. In this way, physicians place themselves in a position to inform patients about issues such as a) why potential health threats are minimal yet require monitoring, b) how deviation from the norm poses health risks, and c) why patients’ should not feel anxiety (see section on reassurance, below) about falling numbers.

Normal as Evidence that Avoids Explicit Labeling of Patients’ Condition

When analyzing diagnostic news deliveries, Maynard (2004) distinguishes between citing the evidence (e.g., reporting test results) and asserting the condition (e.g., making claims about the patient’s overall health). Citing evidence is a more subtle way of communicating diagnostic news, whereas asserting the condition is bold and potentially confrontational. At times physicians preceded asserting the condition by presenting the evidence. On other occasions, they reversed this order. Across other instances physicians cite evidence, like test results, to subtly diagnose the patient rather than overtly diagnosing the patient as sick or well (Maynard, 2004).

Akin to Maynard’s description of the use of test results as evidence in place of overt diagnostic news (e.g., see Excerpt 3, above) the following excerpts reveal that physicians’ use of normal functions in similar ways. When locating patients’ test results, patients’ overall identity as normal or not normal is a cautious resource that avoids labeling the patients’ condition overtly. Rather, from the citation of the evidence, patients infer and respond to their overall condition and health identity. In Excerpt 5 (below), the physician uses normal in reporting the test results and as a means of locating the patient’s results within a preferred range. This assessment can be either binary or gradated (Pillet-Shore, 2003). Here the physician also uses normal (line 4) to clarify a previous biomedical explanation:

Excerpt 5 (OC5:12)

1	Physician:	[And] there’s thirty nine, what’s called seg’s:,=
2	Patient:	=Um hm.=
3	Physician:	And (.) u:m thirty nine lymphs?, thirteen monos, six ↓(eyos), and two (basils).
4	→	So it’s a normal differentia:l. U:h, don’t have to worry about- your
5		(gramatica’s) thirty nine point seven, so you’re not anemic. Your (platelets) are
6		four ↓twenty nine. So there’s nothing there that (0.3) .hh bothers me, or ↓kind of
7		alarms me in any way.

The physician’s listing of test results amounts to reporting good news to the patient. With “Um hm” (line 2) the patient’s minimal response facilitates the physician’s continuation with a biomedical explanation, providing little indication of how the message is received or if it is even understood (Gardner, 2001). The physician’s statement of “So it’s a normal differentia:l” (line 4) explains what the information means for the patient. Here the physician uses the term normal as a binary (i.e., either normal or not normal), rather than gradated metric, to clarify the bottom-line meaning of the test results (Pillet-Shore, 2003). The physician’s “normal differential” announcement previews an explicit statement of “don’t have to “worry” (line 4). These results do not appear to bother or alarm the physician, creating for the patient a way to treat otherwise inaccessible test results as good rather than bad news because they are normal. In this important sense, normal becomes a lens clarifying and tipping the valence of the news being delivered toward good rather than bad information (see Beach, 2009; Maynard, 2003).

Below, the term “uncommon” (line 2) functions in similar ways to the usage of normal in previous excerpts as it indicates the physician’s attempt to provide some reassurance, despite the somewhat concerning bad news of having a lower blood count:

Excerpt 6 (OC6:18)

1	Physician:	.hhh Your blood counts if anything have come do:wn: a little bit,
2		and that’s not too uncommon. .hh When I did your exams today
3		I noticed that if there are a few lymph nodes that are enla:rged,
4		there’s not ma:ny.=
5	Patient:	=Really?=
6	Physician:	→Uhm (.) A::nd (.) the: spleen is of normal siz:e and your liver’s
7	→	of normal size and I can’t feel any abdominal (.) you know, masses
8		in your tummy.=
9	Patient:	Where are the lymphs enlarged?
10	Physician:	Ah, in your neck nodes.=
11	Patient:	=Oh ok[ay.

Good news continues to be delivered as the physician emphasizes that while a few lymph nodes are enlarged “there’s not ma:ny.” (line 4), which elicits a surprised response with the patient’s “Really?”(line 5). In response, the physician twice invokes normal to report that both the spleen and liver are not suspicious (lines 6–7) and, in addition, no masses were felt in the patient’s abdomen (lines 7–8). From these good news announcements, the patient can reasonably conclude that no medical problems exist with the spleen, liver, and abdomen. Additionally, lowered blood tests and few enlarged lymph nodes offer potential good news, disclosed by the physician to minimize heightened threat and enhance overall hope for being well rather than sick. These actions function to counter the earlier concern (line 1) that blood counts “have come do:wn: a little bit” (line 1).

Referring to the Absence of Normal to Indicate Possible Sickness

When managing cancer, test results do not always fall within the normal or preferred range. Instances were identified in which a lack of normal was used by physicians as evidence to indicate that patients faced health risks and may be (more or less) encountering sickness. A deviation from the norm often functions as evidence of a diseased state (Engel, 1977). The following excerpts illustrate the reporting of tests that are not normal. These instances function to deny possession of critical features indicating that patients may not belong to the category or group reflecting wellness (e.g., see Widdicombe, 1998a). Based on the evidence, at times physicians must deny, or simply cannot argue for, patients’ inclusion within categories of normal (and by implication, healthy). Of course, one abnormal test does not necessarily mean the patient is sick; possible bad news can be countered by other hopeful findings (e.g., see Excerpt 6, above; Excerpt 7, below). Nonetheless, each test or risk assessment functions as one possible point of evidence.

The following excerpt involves a 70-year-old patient, who despite two previous and benign biopsies of both breasts, seeks a second opinion from a surgeon about risks and likelihood of being diagnosed with breast cancer. What is notable about this excerpt is the reporting of test results that are “higher than normal” (lines 7 & 9). When factoring in the patient’s age, however, these findings are also treated as “pretty rational numbers” by the physician (line 10):

Excerpt 7 (OC2:5)

1	Physician:	Over five years it says that uh if we took a h- hundred
2		women just like you, five of them or five per cent would get can↑cer.=
3	Patient:	=In the next five years.=
4	Physician:	=In the next five years. And over a lifetime, fourteen per cent=
5	Patient:	=Out of a hundred.=
6	Physician:	=Out of a hundred. <oFourteen out of a hundred.o> Now that’s
7	→	higher than normal by a little bit.=
8	Patient:	=Mm hm.=
9	Physician:	→=This is higher than normal by a f- fair bit, but not considering your age
10		( ). Those are pretty rational numbers. Of course what raises it up is your
11		mom. You can’t change that.
12	Patient:	Right.

Using normal as a baseline and thus preferred range, the physician acknowledges scores that are a “little bit/fair bit” high, including a key feature of patient’s medical history: The patient’s mother died of breast cancer. Despite these potential high risk factors, the interview unfolds by informing and encouraging the patient – especially in light of previous biopsy results and slightly high but not alarming numbers – not to be overly concerned about future diagnosis of breast cancer. Here the physician appears to use normal as a gradated rather than a binary metric (Pillet-Shore, 2003). Rather than assessing the patient as not normal, the physician assesses the patient as “higher than normal by a little bit” (line 7), which is then corrected to “higher than normal by a f- fair bit” (line 9). This indicates a gradient understanding of normal/not normal, where a “fair bit” appears to be more troubling than “a little bit”.

The patient below is informed that the presence of an excess protein in the colon is “not normal” (line 6):

Excerpt 8 (OCD1P10:23)

1	Physician:	So when you see this out here (.) it’s diagnostic of what we
2		refer to as a monomeric protein. There’s the word right there?
3	Patient:	Mm hm.
4	Physician:	Monocamarel (.) that means one colon. Okay? So it means that there’s
5		one type of protein in here that’s in excess and shouldn’t be there.
6	→	It is not normal to be there.
7	Patient:	Right.

Before indirectly delivering what eventuates as bad news, the physician attempts to clarify and define (in lay terms) what “Monocamarel” means (line 4). With the following description of an excessive protein that “shouldn’t be there.” (line 5), the patient is prepared to hear the next “not normal” (line 6) as bad rather than good news. Similar to previous excerpts, the presence of an excessive protein is recruited as evidence to assert a troubling medical condition (Maynard, 2004; Peräkylä, 1998). Once again the physician exhibits the power, as a medical expert possessing epistemic knowledge, not just to label the patient’s symptoms as normal or not normal but to make clear the valence of such an assessment (in this case, some kind of bad news that gets further addressed as this interview unfolds).

Providing Reassurance About Normal and Not Normal

Because cancer is one of the most feared and uncertain diseases (Huler, 2008; McKenzie & Crouch, 2004), physicians provide and patients often make bids for reassurance to mitigate and assuage their fears, and to better understand their own condition. In both cases, physicians may utilize normal and not normal to reassure patients about hopeful possibilities for wellness.

Before examining instances where reassuring actions occur, however, it is important to emphasize that not all normal moments involve reassurances. For example, at times physicians simply use normal as a basic resource for indexing patients’ particular body parts (including cells and blood counts). References to normal often occur in the midst of physicians’ lengthy narratives, efforts designed to educate patients (and co-present family members) about the complexities relevant to their particular medical problems.

For example, prior to Excerpt 9 (below) the physician explains hypothetical treatments for prostate cancer with the patient. The physician then describes how “a few normal prostate cells” (lines 2–3) survive radiation:

Excerpt 9 (ND5P5:21)

1	Physician:	And then because radiation does not remove the prostate the PSA doesn’t go
2		undetectable. It usually goes below one ideally, but there’s a- still a few
3	→	normal prostate cells that survived the radiation. So it levels off at some low
4		level and then you just follow along. ((continues))

Here the physician’s use of normal is clearly offered as a contrast to diseased cells. While the presence of some normal prostate cells offers a glimpse of good news, it is not offered by physician as reassurance but describes the ebbs and flows of PSA levels.

Similarly, the physician in the following instance also uses normal to conclude a lengthy educational narrative. However, in this instance, the physician addresses why the patient’s pap smear came back normal when the patient is, in fact, diseased:

Excerpt 10 (ND14P3:14)

1	Physician:	Um, as you know there’s this cystic mass on your ovaries and that’s, you know
2		about the ultrasound and the MRI, and I do think you will need surgery for that.
3		But I think first we need to work up this mass in your vagina a little bit more
4		before we decide to operate. Um, the original plan was to do the biopsy today,
5		but after examining you, what I feel is that, the mass which is walnut-shaped, um,
6	→	is completely covered by the normal tissue of the vagina. ((continues))

Although the patient’s condition will require surgery, a problem exists because the patient’s mass “is completely covered by the normal tissue of the vagina” (line 6). Here again the reference to normal is not designed to offer reassurance, but to indicate how healthy tissue actually makes it more difficult to access patient’s vaginal mass.

In contrast, when physicians produce activities such as establishing a range of test results and describing how a patient’s condition fits within that range, physicians often followed news of abnormality with reassurance and the framing of possible good news. The following two excerpts exemplify a well-established pattern where bad news is followed by good news (e.g., see: Beach, 2009; Holt, 1993; Jefferson, 1988; Maynard, 2003). Maynard (2003) summarizes how “good news can stand on its own as a transition is made to other topics and activities, but bad news invariably obtains some kind of mediating good-news exit” (p. 183). Thus, the general preference for good news over bad news illustrates how “participants in interaction structure the social world as a relatively benign one” (Maynard, 2003, p. 183). This finding is evident as physicians work to reassure patients that the troubles they face also have positive dimensions.

The patient examined below, who earlier in this interview was found to have enlarged lymph nodes, is considering chemotherapy. The physician below first summarizes the patient’s biomarkers, proteins that “may not be normal now” (line 2), but next states “that’s not necessarily bad” (lines 2–3) and should be checked at a later date:

Excerpt 11 (ND5P3:16)

1	Physician:	Yeah now they- they come down at different rates. The alpha (phi) protein
2	→	(.) may not be normal now, but that’s not necessarily
3		[bad as long as it’s coming down], and then you gotta check again you know later on.

Since the biomarkers come down at different rates, abnormality may be temporary and thus not overly problematic. By explaining the biomarkers as not normal rather than malignant or diseased, the physician presents clear diagnostic news that is positively framed – grounds for promoting realistic hope for the patient rather than unnecessarily increasing fear and uncertainty (see Beach, in press).

The physician below informs the patient, “you probably got some kind of virus” (line 1). During post-transplant, however, the patient is informed of vulnerability to “unusual viruses and viral syndromes” (line 3) such as “shingles” (which the patient does not have):

Excerpt 12 (OC5:12)

1	Physician:	Pt um so, pt I think you probably got some sort of virus, (0.9) cooking
2		in your system. Pt um ↑you are at risk (.) post transplant, you are at risk
3		for unusual viruses and viral syndro:mes. One thing that we do sometimes
4		see is shingles. Okay, now you don’t have shingles. I didn’t see any- you know
5		(.) s- spots on your ski::n. (.) Now this wouldn’t be the typical (program) for shingles.
6		But (.) um this could be an opportunity just to talk about the fact that (.) after a
7		transplant, after oyou knowo a ↑stem cell transplant, for the first ye:ar or so:,
8	→	(.) you’re immune system is not up to normal, [okay.]=
9	Patient:	[(oOkayo)]
10	Physician:	=Um: so that when anything unusual ↑happens, you should call it to my attention.
11		=So you did the right thing.
12	Patient:	oGood.o

These possible problems are eventually explained with an important caveat: It is ordinary that a post-transplant patient’s “immune system is not up to normal” (line 8). The physician’s overview of the test results, and eventual straightforward statement of “not up to normal”, demonstrates how potentially threatening events can be situated as normative occurrences. In essence, the physician establishes medical authority by describing how the patient’s susceptible immune system is a normal outcome that requires monitoring but, and importantly, is not a serious cause for worry.

This assertion can de-trigger potential anxiety, and create a shared orientation where both patient and physician understand that specific symptoms are not uncommon. As Heritage (2009) has noted, the “use of declarative sentences ordinarily establish a positive epistemic gradient between speaker and hearer” (p. 300). A space where what is known about the patient’s condition, and in this case how best to proceed in managing “some sort of virus,” becomes a stance that can be claimed with assurance by both patient and physician. In line 9, the patient quietly affirms the normality of the condition (e.g., see Heath, 1992) followed by the physician further reassuring patient that by scheduling an appointment “you did the right thing” (line 11). In this way, the physician informs patient that the reason for the visit and problem presentation was, indeed, “doctorable” and meriting the attention given to the patient’s concerns (see Beach, 2013; Heritage & Robinson, 2006).

A similar instance occurs in Excerpt 13 (below), as the physician first describes the history of a patient who may potentially have chronic Graft-versus-host-disease (GVHD). This disease is sometimes a side effect of bone marrow transplant and involves the donor cells attacking the body of the host (Johns Hopkins Medicine, 2011). Specifically, chronic GVHD can impact the long-term health of the patient.

Excerpt 13 (OCD2P8:3)

1	Physician:	Right, okay. So um, that’s where things were. Now you did have this
2		episode of quite bad ( ) in December u:m, that may be a feature of
3		chronic GVHD. Sometimes you know, people who have had a ( )
4		transplant sometimes will be susceptible to infection for quite awhile but
5		usually bacterial infections that are serious are something that is a feature
6	→	of chronic GVHD. But ( ) levels have been normal (.) and this was one
7		episode. It’s not like– oh you had some pneumonias in the past as well, correct?
8	Patient:	Right. And I’ve had sinus infections, four to six a year.
9	Physician:	What happened?

The physician first presents possible bad news to the patient, attributing the patient’s symptoms in December (lines 2–3) to a chronic illness, GVHD. However, the physician eventually concludes with “But ( ) levels have been normal (.) and this was one episode” (line 6). This dual-description places patient’s range as “normal”, and further minimizes the threat because it only occurred as a single “episode.” (line 7). The patient’s normal levels, and infrequent episodes, function to counter (at least in part) potential concerns with sickness and provide some good news despite the likelihood of having GVHD (Maynard, 2003). These actions provide the patient with some evidence to be hopeful.

Later in the same interview, a series of six normal references occur in Excerpt 14 (below), all involved in announcing good news and seeking or providing reassurance about patient’s condition. The excerpt begins as the physician summarizes that both “platelet count” and “(IT’s)” are “normal” (line 3). The physician concludes by stating, “they all really look very good.” (lines 3–4):

Excerpt 14 (OCD2P8:11)

1	Physician:	Yeah. You’re ( ) was I think 1.3. (.) Well (I’ll try to find that now).
2		Um maybe it’s in this ( ). Your ( ) is around 1.3, you’re
3	→	platelet count is normal, you’re (IT’s) were normal. So you know they all
4		really look very good.
5	Patient:→	Okay. Is (cranton) – is that at a normal level?
6	Physician:→	Yeah up to 1.5 is normal.
7	Patient:	Okay. So all look pretty good.
8	Physician:	Yeah ( ) can find it. Or if I can’t find it we’ll print it out for you.
9		Okay here it – here it is. Um so this is your um you know – your hemoglobin
10	→	and everything is completely normal. And this was done on 04/20. So it’s
11		not that long ago. I mean you know I wouldn’t be repeating it that often
12		you know or routinely. Um you know your (triglycerites) a little high,
13		Dr. B. can talk to you about that. Your lower function tests were
14	→	essentially normal. ( ) was 123 so this really does not look too bad.
15		He even checked your LDH (.) 163. So that looked good.

On line 5, with “is that at a normal level?”, the patient seeks information about additional test findings and confirmation of possible good news. With the binary assessment “Yeah up to 1.5 is normal” (line 6), the physician provides details confirming normal but does not explicitly state that normal is good news. In response (line 7), patient’s next “So all look pretty good.” solicits further reassurance by echoing the physician’s earlier announcement (lines 3–4). In lines 8–15, the physician not only confirms but provides additional good news, using normal twice to edify the patient’s overall health condition. The physician states, “your hemoglobin and everything is completely normal” (lines 9–10). With “everything” and “completely”, the physician uses extreme case formulations to legitimate the assessment and “to assert the strongest case in anticipation of non-sympathetic hearings” (Pomerantz, 1986, p. 227). This utterance is designed by the physician to ward off the patient’s anxieties, fears, or doubts that may keep the patient from not believing the strength of the claim that the physician advances. Essentially, the stance the physician constructs upgrades normal by offering (with “everything is completely normal”) an even more emphasized and positive assessment of the patient’s normal condition.

Discussion

Though largely overlooked, normal is a key element during medical interactions and especially throughout the diagnosis and treatment of cancer. During oncology interviews normal is found to be a surprisingly prevalent term, recruited to achieve a critical set of social actions comprising what is perhaps the most primal aspect of patient-provider interactions: Determining, assessing, and treating patients’ health status as either well and/or sick. Although the word normal appeared frequently throughout oncology interviews, little if any research has examined the important sequential environments surrounding these key moments. A systematic analysis of normal begins to reveal how social conduct during oncology interviews strikes at the very heart of wellness/sickness experiences, and by implication, the normal human tendency to seek courses of action toward being healthy rather than diseased.

This analysis focused on how physicians employ normal to accomplish an array of important social actions that are central to cancer care: Invoking normal as a preferred range, recruiting normal as evidence without explicitly informing patients of their condition, treating the absence of normal as indicative of sickness, and countering potentially bad with good news by reassuring patients about normal and not normal depictions of their condition. Our findings reveal a series of implications for future research.

First, the study only begins to explicate the relation of identified normal social actions to the overarching management of talk about wellness and sickness. Although physicians rarely used terms such as healthy or sick, physicians frequently invoked normal or not normal in the discussions of tests, symptoms, and body parts to subtly inform patients about their health conditions and to identify patients as (more or less) well and/or sick. While normal was the overt invocation, labeling and identifying implications for being well and sick are primary goals for cancer care. This raises key questions about how physicians directly and indirectly inform cancer patients about their bottom-line health status. It also suggests that in the presence of normal/not normal attributions, patients are provided with opportunities to both clarify and negotiate their condition with physicians. However, in the data excerpts examined herein, patients and family members displayed relative passivity when hearing normal/not normal descriptions by physicians, responses which have been found to be quite normal in the face of diagnostic news (e.g., see Heath, 1992).

Second, the insights gained from this study can further advance understandings of medical encounters by making clear how not only physicians invoke normal, but how patients also use and rely on normal/not normal to describe and explain their lay understandings about diagnosis, treatment, and prognosis. A collection of these patient-initiated moments have been identified, are currently being examined, and will provide an important contrast for laying bare an identified and recurrent tendency to strive for normality as wellness. Analysis of these instances will further elucidate how normal is used by patients, not only whenever patients seek physicians for healing within clinical environments, but also in homes and more casual settings as patients and family members attempt to make sense of, and as best possible deal with, ongoing cancer journeys (e.g., see Beach, 2009).

Third, related but unexplicated issues of gender and ethnicity were also not specifically addressed in the present analysis. However, previous research indicates that female primary-care physicians used more patient-centered communication and had longer interview times than male primary-care physicians, suggesting gender differences in medical interaction (Roter & Hall, 2004). Given these findings, it would be interesting to explore if female physicians systematically used normal differently than their male colleagues. Additionally, because communication is influenced by culture (e.g., see Giri, 2006), it would also be useful to explore how or if alternative orientations to normal are based on varying ethnicities, enacted by diverse populations in culturally specific ways.

Fourth, the data analyzed herein make clear that physicians’ ability to describe and classify normal and not normal circumstances embodies a set of fundamental practices for establishing their expertise and authority as medical professionals (e.g., see Heritage, 2009; Peräkyla, 1998, 2002). Physicians rely on their exclusive knowledge and expertise to discern and report preferred ranges of health, and thus place themselves in a powerful position to ultimately deliver news about the patient’s bad/good condition. Further attention needs to be given to physicians’ abilities to accurately summarize, update, and attempt to educate patients about tests and results (e.g., see Venetis, Robinson, & Kearney, in press; Robinson, Hoover, & Venetis et al., 2013). The potential for promoting shared understandings, alleviating anxiety and suffering, and facilitating comfort and healing in the midst of cancer care is a primary task and challenge for future research endeavors. Additionally, even when patients receive good news, on what occasions do they align or even resist by focusing on bad news scenarios (e.g., see Beach, 2013d; Drew, 2013; Frankel, 2013)? Future research and clinical practice could benefit by more closely examining how patients do or do not subordinate to physicians’ normal/not normal classifications, and on certain occasions, not only resist but actively challenge physicians’ assessments.

Finally, only minimal attention was given to non-overt uses of normal, like common and ordinary. For example, in Excerpt 6 (above) as the physician comments to the patient, “Your blood counts if anything have come do:wn: a little bit (.) and that’s not too uncommon.” the physician attempts to reassure the patient of inclusion within the preferred range. This range appears functionally similar to normal, but a more precise understanding of alternative practices for describing and categorizing test results, and their consequences for sustained health, would provide a more detailed understanding of how wellness and sickness get socially constructed during oncology and other interviews.