Strategies in Medical Interviews: Patients' Explanations & Doctors' Responses, Study Guides, Projects, Research of Sociological Theories

Download Strategies in Medical Interviews: Patients' Explanations & Doctors' Responses and more Study Guides, Projects, Research Sociological Theories in PDF only on Docsity! 5 Explaining illness: patients’ proposals and physicians’ responses Virginia Gill and Douglas W. Maynard Introduction Patients visit doctors for a variety of reasons, and a prominent one is to find out what is causing some health problem or symptom they are experiencing (Korsch et al. 1968; Novack 1995). However, during the course of medical interviews, patients often offer their own “lay” or “folk” explanations for what is causing their health difficulties. In the view of many researchers, doctors routinely ignore or dismiss patients’ theories (Cicourel 1983; Fisher and Groce 1990; Kleinman et al. 1978; Mehan 1990; Mishler 1984; Stoeckle and Barsky 1981; Waitzkin 1979, 1991). Doctors, because of their power and author- ity, are said to impose a biomedical perspective upon patients.1 They maintain an exclusive focus on only those symptoms and disease processes that are under the purview of the medical model, rather than considering or appreciating patients’ social experiences and perspectives regarding their illnesses. Thus, the “voice of medicine” regularly silences the “voice of the life world” (Fisher and Groce 1990; Mishler 1984). Despite such pronouncements, investigators have not described or analyzed, in any detail, the interactional structure of patients’ explanations and doctors’ responses (explanation– response sequences) as they occur within the context of clinic visits. Nor has enough attention been given to how explanation–response sequences relate to ongoing courses of activity, such as the different 1 See the review and critique of literature on the “asymmetry” in doctor–patient interaction, in Maynard (1991c); Robinson (2001a). 115 116 Virginia Gill and Douglas W. Maynard phases of the medical interview.2 This chapter draws on audio- and videotaped data of patient visits to an outpatient medical clinic3 to examine the interactional strategies patients use to offer expla- nations for their medical problems and the methods doctors use to respond to these explanations. Our focus is how patients design and place their explanations in the phase of the medical interview where doctors are gathering information about symptoms. What has been portrayed as a struggle between the doctor’s “biomedical” perspective and the patient’s “lifeworld” concerns can be recharacterized in terms of interactional dilemmas that doctors and patients face. These dilemmas involve sequential organization within two ordered phases of the medical interview: (1) the col- lection of medical data through verbal and physical examination or Byrne and Long’s (1976) phase III of the interview; and (2) the “con- sideration” or analysis of this data or Byrne and Long’s (1976) phase IV. During clinic visits, patients show that they face this dilemma: How, within a course of action that primarily involves collecting medical data (facts about the nature of patients’ symptoms and other aspects of their experiences) can patients offer their analyses of these facts (explanations) so that doctors may consider them, yet with- out requiring such consideration immediately, in the data-gathering context?4 That is, when the doctor is gathering facts about a par- ticular symptom, it provides an opportunity for the patient to offer 2 Conversation-analytic and ethnographic researchers have considered patients’ or “lay” perspectives when investigating other topics in medical or clinical interac- tions (Drew 1991; Heath 1992; Heritage and Sefi 1992; Maynard 1991c, 1991d; Silverman 1987; Stivers 2002b; Strong 1979; ten Have 1991), but few have given primary attention to patients’ actual explanatory practices and doctors’ responses (see Gill 1995, 1998a, 1998b; Gill et al., 2001; Raevaara 1998). 3 Data were collected by the second author at an outpatient general internal medicine clinic associated with a teaching hospital in a medium-sized city in the Midwest- ern United States. The data corpus includes 15 audio- and videotapes of clinic visits (involving 15 patients and 5 physicians), and 2 audiotaped follow-up calls (involving 2 of the patients and 2 of the doctors). 4 A related dilemma is discussed in Gill (1998a): patients often have explanations for their illnesses, but treat as problematic any display of personal authority about these explanations. Patients handle the dilemma by displaying certainty about the explanations in contexts where they are not inviting doctors’ assessments. Con- versely, patients downplay their certainty when their explanations solicit evalua- tion from doctors. Thus, patients do manage to insert their explanations into the medical interview yet refrain from requiring doctors to recognize them as author- itative sources of this type of knowledge. See also ten Have (1991), who argues that appearing “uncertain” is a way for patients to put explanations on the table yet maintain a subordinate role vis-à-vis the doctor. Explaining illness 119 symptom. In excerpt (2), Ms. O uses “because” to attribute her depression and upset feelings to lack of sleep. (2) [2:104] 1 Ms. O: Well as I said I think I get: (.) tah: (2.0) depressed and 2 → upset because I can’t (0.8) I’m not- getting sleep. Patients also use forms of the indicative, such as “is” or “was,” as attributive linkage proposals. The causal factor in these cases is a hypothetical bodily condition. In excerpt (3) below, Ms. N uses “was” to question whether her rectal pain can be attributed to the condition, “hemorrhoids.” (3) [10:594] 1 Ms. N: While we’re on my gut. 2 Dr. D: ◦Yes.◦ 3 Ms. N: A couple a weeks ago: hh u:m (0.6) I had (.) tremendous 4 amount of rectal pai:n? 5 (1.0) 6 Ms. N: → No:w- whether it was hemorrhoids or not I’m not sur:e 7 because there was a lot of (0.8) .h pai::n when I tried (0.2) pressin:g. Similarly, patients may propose that a hypothetical condition “brought on” the complaint. In excerpt (4), Ms. A and Dr. A are discussing the patient’s chest pain: (4) [6:383] 1 Dr. A: An so tha:t (.) came on with the exerci:[se 2 Ms. A: [M hm? 3 Dr. A: An- with other activities that you’ve do[ne. 4 Ms. A: [M hm? 5 Dr. A: ◦Oka:y:◦ 6 (0.5) 7 Dr. A: .hh (2.5) An in addition sometimes you wake at nigh:(.)[t wi]th that. 8 Ms. A: [M hm] 9 (3.5) 10 Ms. A: → An I was wondering if: ◦you know◦ stress could a (.) brought that on 11 too. Patients may link a pain or other symptom to a specific site of origin, such as an organ in the body, by proposing that the symptom is “in” that organ. Below, Ms. B cites her “gall bladder” 120 Virginia Gill and Douglas W. Maynard and then her “kidney” as the cause of the abdominal tenderness she is experiencing: (5) [7:365] 1 Ms. B: .hhh An: : :d then I get a lot of tenderness: in this area hh. 2 And again:, it’s probably: (1.0) [whether: it’s] in the = 3 Dr. A: [In the front] 4 Ms. B: → = gall bladder? Kidney? Thus, patients’ overt explanations are based on a three-part turn structure, consisting of a complaint (reference to a symptom or other discomfiting health problem); an attributive linkage proposal; and a causal factor – a reported circumstance, hypothetical bodily condi- tion, or site of origin. Patients put these elements together to account for the existence of their symptoms. Tacit explanations A patient can offer a tacit explanation by describing or referring to a symptom and then reporting a life circumstance or experience.8 The patient connects these elements with a non-attributive linkage proposal such as “and” or “but.”9 The patient invites the doctor 8 Reporting is a generic strategy that speakers can use to accomplish various types of tacit or implicit activities in conversation. For example, a speaker can avoid taking an official position in relation to a proposal, such as an invitation, by producing a report of an activity or circumstance in response (Drew 1984:134): I: How about the following weekend. (0.8) C: → .hh Dats the vacation isn’t it? I: .hhhhh Oh:. ALright so:– no ha:ssle, . . . . In this excerpt, I issues an invitation. C’s subsequent report (arrowed) provides I with “the materials from which she can see for herself that it will not be possible to go then” (Drew 1984:134). However, C leaves it to I to determine the implication of his report. I takes the report as a rejection of her proposal. 9 Whereas “and” projects that a forthcoming utterance is “additional” (in relation to a previous utterance) and thereby proposes a connection between the two, “but” can be used to propose a relationship between two utterances by setting off a forthcoming utterance against a prior utterance. A variation is for patients to use “since” to propose a temporal relationship between a symptom and another circumstance, and thus tacitly suggest a causal connection. See Drew and Heritage’s (1992:31–2) example, taken from Mishler (1984:165): Dr: How long have you been drinking that heavily? Pt: Since I’ve been married Dr: How long is that? Pt: (giggle) Four years Explaining illness 121 to analyze the report’s relationship to the complaint, while stopping short of overtly proposing that those circumstances or experiences are causal factors. The patient merely implies or hints at such a rela- tionship and provides the doctor with the opportunity to display recognition of the “upshot” and to officially make a causal con- nection between the patient’s report and the patient’s symptom (see Drew 1984; Gill 1995, 1998b; Strong 1979). For example, in excerpt (6) Ms. B offers a tacit explanation. She complains of a symptom (line 1) and then reports that she has a new car (lines 1 and 3). The doctor does not immediately display recognition of an upshot. After two more reports related to the emergence of her symptom (line 6) and time spent sitting in the car (“ = ”shifts,” line 7), the patient herself goes on to propose (in a speculative manner) a causal connection between the car and the backache (line 9): (6) [10:523] 1 Ms.B: .hhh I’ve been having this backache:KHH. .hhh A[n:d we ] do = 2 Dr.A: [Do you:] 3 Ms. B: = have a new car:::, 4 (1.0) 5 Dr.A: ◦M [hm◦ 6 Ms.B: [An:::d (.) it- (.) didn’t bother me the first two weeks. 7 But we did do: a couple of three hour:: shi[fts. 8 Dr.A: [Mm hm? 9 Ms.B: Whether that’s it:thh? 10 (0.4) In contrast, in excerpt (7) the patient issues a tacit explanation and the doctor’s response does immediately display recognition of an upshot. That is, by proposing to look for “underlying causes” for the patient’s fatigue (lines 8 and 11), he also shows his understanding that in lines 3–5 the patient was offering an explanation for her fatigue, and that the real cause may be more serious than “burning the candle at both ends”: (7) [16:1032] 1 Dr. C: You mention some easy bruising? An bleeding? Fatigue? 2 Ms.I: Uh. I- an the- an: Ut you know: has been (.) Ut recently 3 that I have the fatigue. But I guess: you know: you’re just 124 Virginia Gill and Douglas W. Maynard Explanations that do not strongly compel doctors’ confirming or disconfirming assessments By design, a tacit explanation – complaining about a symptom and then reporting a fact or circumstance – puts very little pressure on the doctor to respond with a confirming or disconfirming evalua- tion, as in excerpts (6)–(8); that is, this set of actions does not firmly initiate an explanation–assessment sequence. The patient gives the doctor the option to display recognition of an upshot, but also gives the doctor the option to hear the report as simply that – a report of circumstances. The doctor may relevantly treat the report as “infor- mation” or “data” and proceed with information-gathering activi- ties by simply nodding, or otherwise indicating receipt of the report. To say that the physician can relevantly take this option does not mean that this is the best option. It simply means that the patient does not put the doctor in a position where he or she must respond, or else appear to be ignoring an explanation that the patient put on the table. Officially, there is no explanation on the table for the doctor to evaluate. Even though they officially propose causal connections, the design of some overt explanations can also put little pressure on doctors to produce an assessment. Patients often pose their overt explanations as speculations or out-loud musings,10 which not only display uncertainty, but are also relatively non-constraining in terms of the responses they require from doctors (Gill 1998a); see for example, excerpts (3) to (5) as well as line 9 in (6). Speculative expla- nation are not forthright questions, and therefore do not clearly constitute the first part of a question–answer adjacency pair. If such explanations did, then that would firmly establish the “conditional relevance” of a doctor’s confirming or disconfirming assessment, such that it would be “noticeably absent” were it missing (Schegloff 1972:76–77; Gill 1998a; ten Have 1991). Instead, speculative expla- nations provide for the relevance of an array of responses. Similarly, overt explanations designed as qualified proposals, as in excerpt (2), are also relatively non-constraining for the doctor. Like a “first assessment” (Pomerantz 1984a:61), a qualified pro- posal makes a confirming or disconfirming assessment relevant as a next-turn activity, but does not require such a response. A distinctive 10 See Sacks (1992b:405) on “musing aloud.” Explaining illness 125 feature of qualified proposals is that they display slightly more cer- tainty than do speculative explanations (patients preface them with “I think” rather than “I’m wondering if” or “I don’t know if”), but, as we will show, patients offset this certainty by using quali- fied proposals in comparatively low-risk contexts, where agreeing assessments are likely – for example, when patients are proposing an explanation that is in line with the doctor’s own displayed view. Patients’positioning of overt explanations within turns can also lessen the degree to which they compel doctors’ assessments. Patients may place overt explanations within multiple-component turns that include both symptom-related and explanation-related components. This gives doctors the option to relevantly respond to either com- ponent (Gill 1998a). For example, a patient may construct a two- component turn – reply + explanation – wherein the patient replies to a doctor’s symptom-related question (regarding when a symptom occurs, how long it lasts, etc.) and then offers an overt explanation for the symptom. As Frankel (1990:237) notes, this type of turn design provides the doctor with “an option rather than an obliga- tion” to respond to the second turn component (the explanation). This design is evident in excerpt (9) below. Dr. C seeks a confirma- tion of a problem Ms. I had mentioned earlier in her clinic visit, that she experiences pain with intercourse (lines 1–2). (Several years before, Ms. I had a surgery that included both a hysterectomy and bladder repair.) (9) [19:1259] 1 Dr.C: .hh Kay. An then the other- the other thing you mentioned 2 was (.) you have (.) pain with intercourse. Is that right? 3 Ms.I: Yeah. But that’s just since I’ve had that hysterectomy. An I 4 don’t know if that bladder tie up? Was part of that? 5 (0.8) 6 Dr.C: For th last six or ten years. Ever since that [surgery. So] 7 Ms.I: [M hm? M] 8 hm? Ms. I replies (“Yeah,” line 3), clarifies the date of the onset of the pain (“since I’ve had that hysterectomy,” lines 3–4), which may also tacitly suggest that the surgery is the cause of the pain, and then she adds a more overt, speculative explanation concerning a “bladder tie up” (line 4). This turn design presents Dr. C with the option of 126 Virginia Gill and Douglas W. Maynard focusing on either the reported timing of the patient’s pain, or her explanation. He responds in terms of the timing of the pain (line 6) and goes on to query about its frequency (not in excerpt). Patients may also add other turn components to overt explana- tions, so that the explanations themselves become less assessment- relevant. For example, patients may append turn components that return the talk to the activity of describing their symptoms. Then the doctor may attend to the descriptive portion of the patient’s turn. For example, in excerpt (10), Dr. B asks Ms. D whether she experiences shortness of breath (line 1). Ms. D replies to the query with “So:me” (line 3), then produces her explanation (“that-’s:?cuz I should lose wei:ght.”). She adds a tag component: a downgraded description of how much shortness of breath she experiences (“NOT much,” line 6): (10) [9:431] 1 Dr.B: Shortness of brea:th? 2 (1.0) 3 Ms.D: So:me: but that-’:cuz I should lose wei:ght. I know 4 that. 5 (.) 6 Ms.D: I think- NOT much. 7 Dr.B: When do you get short of brea:th. Dr. B’s next query (line 7) is directed toward her temporal experience of the symptom, rather than her explanation. Patients consistently use this turn design, which “envelops” the explanation within turn elements that describe a symptom or cir- cumstance, when they offer unmitigated overt explanations, as in Ms. D’s line 3 utterance above (Gill 1998a). Accordingly, the expla- nations that patients deliver with the most certainty do not actively solicit assessment. We shall return to this matter later, when we explore other kinds of work that unmitigated overt explanations can do. Explanations that strongly compel doctors’ confirming or disconfirming assessments Patients may pose their overt explanations as frank questions that narrowly restrict doctors’ response options, such that doctors are Explaining illness 129 Ms. A speculates whether the hypothetical condition, “hormone deficiencies,” could cause dry skin (lines 1–4). (12) [9:539] 1 Ms.A: The only thing I was wondering if dere is .hhhhh you kno:w 2 ah::n (2.0) ((doctor turns from desk to look at patient)) 3 hormone deficiencies or something like this that it (0.6) 4 (>◦you know◦<) that dries your skin out too. 5 (0.5) 6 Dr.A: ◦Mm◦ 7 (0.5) 8 Dr.A: Tch .hhh ah:m 9 (0.8) 10 Ms.A: Or no[t too much 11 Dr.A: [tch There are some hormone problems like thyroid 12 p[roblems] = 13 Ms.A: [◦Mm hm◦] ((nodding)) 14 Dr.A: = which can do tha:t. Um we’ve never found that (.) on you 15 before. 16 Ms.A: No = ((shakes head)) 17 Dr.A: = (though) we could certainly think about- ◦about that.◦ 18 Ms.A: An- how did my cholesterol test turn out. 19 (.) 20 Ms.A: Blood tests I’m curious about tha:t. At lines 5–9, Dr. A delays answering and produces tokens that may portend disagreement, whereupon Ms. A revises her explanation in a way that anticipates a negative answer (line 10).12 In partial over- lap with Ms. A’s revision, the doctor offers a disconfirming assess- ment (lines 11–12 and 14–15). She claims that, in Ms. A’s case, there is insufficient empirical evidence to support the explanation. Dr. A’s offering is cautious, in ways that “dispreferred” responses are canonically performed (Pomerantz 1984a). In addition to her initial delays (lines 5–9), she confirms the theoretical possibility of such an explanation (lines 11–12 and 14). But then she “cites the evidence” (Maynard 2004) in a way that could rule out these hor- mone problems in Ms. A’s case (lines 14–15). Note that by referring to evidence from previous lab tests (“Um we’ve never found that (.) on you before”), she displays still more caution in that she does not rule out the possibility that the patient may currently have such 12 Ms. A is revising in a way that observes the preference for agreement (Sacks 1987). 130 Virginia Gill and Douglas W. Maynard hormone problems. Ms. A shakes her head and says, “No,” display- ing knowledge of these findings (line 16). Dr. A goes on to qualify her disconfirmation, portraying herself as still willing to consider the matter (line 17). However, Ms. A does not pursue it any further. She shifts the topic, inquiring about the results of her recent cholesterol and blood tests (lines 18 and 20). Doctors’ immediate confirming assessments While it is more common in these data for doctors to provide confirming than disconfirming assessments in response to patients’ explanations, confirming assessments also occur relatively infre- quently. As mentioned, there are 9 cases out 63 overt explana- tions where, without first initiating an extended series of symptom- related queries and responses and/or a physical examination, doctors respond with confirming assessments after patients offer their overt explanations. Not surprisingly, doctors tend to give confirming assessments in response to patients’ explanations that have exhibited alignment with doctors’ displayed perspectives. Even so, doctors’ confirma- tions are cautious rather than forthright. Excerpt (13) shows a doc- tor giving a qualified confirming assessment in response to a patient’s explanation. Mr. E has pain in his forearm that Dr. B has provision- ally diagnosed as being caused by ulnar nerve entrapment syndrome (a pinched nerve in his elbow). This clinic visit is a follow-up visit; the doctor is evaluating the patient’s condition since he began using an elbow pad and an anti-inflammatory drug. Dr. B first examines Mr. E to determine whether he is developing muscle weakness in the afflicted arm. He explains that activities that involve vibration, gripping tight, and holding the arms bent for an extended period will be irritating (lines 1–2, 4–5, and 7). (13) [5:258] 1 Dr.B: It makes sense that things like mowin’ thuh lawn cuz ya know 2 you’re grippin’ tight an’ [your arms are bent] an’ you’re = 3 Mr.E: [( that’s right] 4 Dr.B: = holdin that position for uh long time ‘n there’s vibration, 5 n’ that’s all [irritating. So it] makes sense that those 6 Mr.E: [(That’s right) ] 7 Dr.B: = kinds o’ things’re gonna bother it. 8 (.) Explaining illness 131 9 Dr.B: .h[h 10 Mr.E: [.hh tch .h I think that >what it is < um- (.) da:maging 11 wa:s:a- (.) I- I do:n’t (.) remember whether I me:ntioned to 12 you or not:a- was years ago:, almost like a- (.) bout six, 13 se:ven years ago. .hhh I work in a workshop in this machin:e 14 jis those:a- those gu::ns? Needle sh:ape 15 Dr.B: Yeah = 16 Mr.E: = to- drill da h:ole ta- 17 Dr.B: Yea[h 18 Mr.E: [ta (glue). 19 Dr.B: Right. 20 Mr.E: And those one I probably work on em (.) constantly work on 21 one ti:me. I[for]go:t, 22 Dr.B: [Mhm ] 23 Mr.E: I didn’t (.) [pay ] attention. 24 Dr.B: [Okay.] 25 Mr.E: And I cont:inuously (0.5) sh:ake it. 26 Dr.B: Righ[t 27 Mr.E: [>I feel< that’s what’s the damaging 28 Dr.B: Yeah = 29 Mr.E: = ◦Ya◦ [(cause for that)] 30 Dr.B: [It may ha:ve. ] 31 Dr.B: Yeah. Mr. E agrees (line 6) and further responds (lines 10–14, 16, and 18) by offering a qualified explanation for what may have initially caused the damage to his arm: he worked at a machine that “contin- uously” shook his arm. Dr. B marks that he is following the patient’s narrative by offering continuers and other tokens of acknowledg- ment, including indications of agreement (lines 15, 17, 19, etc.), even as the narrative progresses (lines 20–21, 23, and 25). A summarizing turn (lines 27 and 29) refers back to the circumstances he reported in his story and the condition he experiences, more overtly proposing that the circumstances caused the damage. Dr. B offers an agree- ment token (line 28), and qualified confirming assessment, “It may ha:ve” (line 30). Thus, in a context where the doctor has already gathered information and made a candidate diagnosis (a pinched nerve), and where the patient’s explanation for what caused the problem (shaking the arm) is in line with this diagnosis and the doc- tor’s pronouncement of what could irritate the arm (vibration), the doctor produces guarded agreement. And Dr. B produces it quickly (line 28) – in the way that preferred responses are done – after Mr. E finishes his explanation. 134 Jeffrey Robinson (1984:120) has observed, doctors do not show the reasoning that underlies their queries.15 However, this focus on gathering empirical data is not a unilateral accomplishment, nor is it simply a matter of the biomedical model suppressing “lifeworld” concerns. Although Mishler (1984:115) contends that patients may be “confused by shifts in the content of the physician’s questions” and have “no clear idea of what [the physician] is trying to discover,” our data show a more bilateral ori- entation toward the activity that predominates in the information- gathering phases of the interview: gathering medical data. Even if patients are unfamiliar with the exact diagnostic agenda physicians may be working to establish through their queries, this should not imply either naivety or passive acceptance of the biomedical model. Instead, patients display an understanding of the interactional struc- ture of the medical interview and the activities through which the biomedical model is realized. When patients place their own analyses within the data-gathering phase of the medical interview, they design and position these expla- nations in ways that accommodate continued investigation or fact- finding. That is, as we observed earlier, where doctors are collecting data about patients’ physical states, patients’ explanations are not sequentially restrictive; they do not constrain doctors to produce confirming or disconfirming assessments then and there. Patients’ strategies for offering explanations thus adroitly handle the inter- actional dilemma noted in the introduction to this chapter: They allow patients to put explanations on the table for doctors’ consid- eration, without being seen to request an assessment “prematurely,” before all the facts are in. For their part, doctors capitalize upon the non-restrictive design and placement of patients’ explanations and respond in ways that focus on what patients are experiencing rather than on why they are experiencing it. In addition, there are cases where patients’ explanations invite rather than merely allow responses that focus on what they are experiencing. For example, patients may use explanation formats as vehicles to introduce and draw doctors’ attention to additional concerns or complaints that may otherwise be difficult to put on 15 Similarly, ten Have (1991:150) observes that in other positions (such as in the third turn position) physicians also “refrain from commentary, utterances displaying alignment, or any indication of their own information processing.” Explaining illness 135 the table. In the next section, we show extended explanation– query sequences, highlighting how doctors delay or avoid producing immediate confirming or disconfirming assessments in information- gathering phases of medical visits, while pursuing their information- gathering activities. First, we show how doctors’ focus on what is occurring (symptoms) rather than why it is occurring (explanations) leads to patients’ explanations being disattended for the entire inter- view. Subsequently, we examine how doctors may eventually con- firm patients’ explanations, after extended insertion sequences that deal with the nature of the patients’ symptoms. Query focuses on the patient’s symptom, no assessment occurs In the next excerpt, a patient offers an explanation that never receives a confirming or disconfirming assessment from the doc- tor. Dr. B has been taking Ms. D’s health history.16 At this point in the interview, he is gathering information about a variety of mat- ters such as her family members’ health, whether she smokes, and whether she experiences headaches or asthma. He then asks Ms. D if she experiences shortness of breath (line 1 below). She replies that she does have some shortness of breath, and then offers her weight as a cause for this condition (line 3). This reference ties back to the beginning of this medical visit, where they had discussed the fact that Ms. D had gained eleven pounds since her last appointment, despite the fact that she had said she was going to try to lose weight. In a laughing way, Ms. D displayed incredulity about this situation, and doctor and patient joked back and forth about what would have caused the weight gain. Her explanation at line 3 may be a way for her to display some authentic concern about this weight gain. How- ever, Dr. B, at line 7, focuses away from the weight gain and on the shortness-of-breath symptom: (14) [9:431] 1 Dr. B: Shortness of brea:th? 2 (1.0) 3 Ms. D: So::me, but that-’s cuz I should lose wei:ght. I know 4 that. 16 This interview, a portion of which is in excerpt (14), is explored in the chapter by Boyd and Heritage (this volume). Also see Gill’s (1998a) analysis of the patient’s portrayal of knowledge in this excerpt. 136 Jeffrey Robinson 5 (.) 6 Ms. D: I think- NOT much. 7 Dr. B: When do you get short of brea:th. 8 (1.0) 9 Dr. B: Stair:s? An: nat sort a thing? Er 10 (1.0) 11 Ms. D: We:::h >after about-< (.) three fli:ght:s or four. 12 HIH huh huh huh. 13 . (1.5) 14 Ms. D: .h ◦Two.◦ N(h)o. Huh. .hhh 15 (1.5) 16 Ms. D: Rea::lly not- not much. Uh uh. 17 Dr. B: ◦Okay.◦ 18 (7.0) 19 Dr. B: Are your bowel movements normal? We noted earlier that Ms. D’s response (lines 3–4 and 6) allows the doctor to focus on gathering information rather than requir- ing him to assess immediately whether her explanation is correct. At line 6, “NOT much” seems to characterize her shortness of breath rather than the weight she needs to lose; i.e., it appears to be a downgraded recharacterization of her initial reply (“So::me,” line 3). Thus, her explanation by this point is effectively enveloped between two descriptions of her symptoms. Dr. B queries her about the shortness of breath, asking her to specify when she experiences this symptom (line 7). This query does not mark whether (or how) it is related to her explanation. After a silence (line 8), Dr. B produces a candidate answer (line 9) for the patient to confirm or discon- firm. Ms. D gives a characterization of how many flights of stairs it takes for her to become short of breath (lines 11–12), and appends a laugh. Next (line 14), she very softly upgrades this to “two” flights (i.e., she produces a characterization that displays the condition as more serious) and then quickly disclaims this upgrade (“N(h)o”), adding more laugh tokens and reasserting her line 6 recharacteri- zation of her symptom: “Rea::lly not- not much. Uh uh” (line 16). As the laughter may be a display of “troubles resistance” (Jefferson 1984b:351), Dr. B (line 17) appears to accept it, and moves on (line 19) to another query related to Ms. D’s health history. Thus, when Ms. D inserts her explanation into an information- gathering phase of the medical visit, she provides for the doctor to Explaining illness 139 her about this possible causal factor; she asks the patient if she is experiencing stress (line 14). Ms. A takes the doctor’s query as an invitation to describe the problems she is having with her teenage son (lines 16–17). Dr. A displays familiarity with these problems (line 18), at least insofar as they have been “mentioned” before.18 However, her claim of prior knowledge does not invite explication of the trouble; that is, as Jefferson (1988:425) has argued, there are two types of responses to a troubles announcement: one which marks arrival [at a troubles telling point] and elicits further talk on the matter but does not necessarily align recipient as a troubles- recipient . . . and one which, by displaying “empathy,” commits recipient as, now, a troubles recipient. Ms. A treats Dr. A’s claim of prior knowledge as less than empa- thetic. In her “further talk on the matter,” Ms. A plays down the problem’s effect on her (line 19), shows a kind of resigned attitude (line 21), and claims that she has a remedy for the problem (lines 21–23 and 25). Ms. A thereby displays what Jefferson (1984b:351) calls “troubles resistance.” Still, after writing in the patient’s file (line 26), Dr. A re-topicalizes the problem by making a candidate assertion about where the son is living (line 27), which invites Ms. A to provide more information (line 28). Dr. A responds to the announcement that Ms. A’s son is “staying with his girlfriend” with a kind of news receipt (line 29) that again discourages troubles-talk development (Heritage 1984b; Jefferson 1981a; Maynard 1997). At line 30, Ms. A confirms that receipt and then (line 32) offers a fur- ther aspect of the trouble, after which there is nearly a half-minute silence during which Dr. A is writing in and examining the patient’s record. Subsequently, Dr. A invites Ms. A to bring up “other things” she may want to discuss (lines 34–36), whereupon Ms. A asks the doctor a question about her skin (lines 39–40), and they continue on that topic. The issue of whether stress could cause her chest pains does not get resolved here, nor later in the interview. Turning to the issue of how Ms. A formulates her explanation in the first place, note that she offers “stress” as a causal factor 18 There is no previous mention of the problems with her son in this interview and we take it that the physician is recalling something from a previous visit. 140 Jeffrey Robinson without also asserting that she is experiencing stress. This resem- bles a device Sacks ((1992b:405) has described, wherein a speaker “muses aloud” using an abstract reference to elicit interest in what he or she “meant by that.” By implying but not actually asserting that she is under stress, Ms. A may be inviting inquiry about whether she is, in fact, experiencing stress. As noted, the doctor’s empirically focused response (“◦Are you feelin: stressed?◦”, line 14) gives her an opportunity to talk about the stress she is under. While Ms. A appears to make attempts at troubles talk and to discuss what the medical literature calls psychosocial issues (Engel 1997; Stoeckle 1995; Frankel et al. 2003), Dr. A does not respond further in a way that encourages such talk. Nevertheless, we can see how a patient may use the doctor’s orientation toward gather- ing medical data to occasion the delivery of announcements about troubles, psychosocial issues, or other matters that would not oth- erwise easily fit within a context where the physician is gathering information about a particular symptom. Query focuses on the patient’s symptom, assessment occurs after a delay As noted, the non-constraining design of patients’ explanations allows doctors to focus away from the explanations and onto patients’ symptoms. Eventually, doctors may assess patients’ expla- nations. This happens in the following excerpts. After the patient offers an explanation for a symptom, in excerpt (16a), the doctor initiates an extended series of symptom-related queries and con- ducts a physical examination. She eventually confirms the patient’s explanation in excerpt (16b). Thus, the queries, replies, and exami- nation become an extended insertion sequence between the patient’s explanation and the doctor’s eventual assessment.19 In excerpt (16a), Dr. A and Ms. B are in the phase of the medical interview before the physical exam where the patient is introduc- ing her physical symptoms. She is holding a piece of paper, which she looks at as she reports experiencing tenderness in her abdom- inal area (line 1). As she raises this symptom, she also indicates its location with a gesture, motioning under her right rib. Then 19 Similarly, Whalen et al. (1988) discuss an “interrogative series” that operates like an insertion sequence between a caller’s request for help and a call taker’s announcement of dispatch in 911 calls. Explaining illness 141 Ms. B appears to start a relatively firm explanation (“It’s probably:”, line 2), which she abandons in favor of a speculation as to “whether” the discomfort is “in the gall bladder?” (lines 2 and 4). Ms. B thus produces an explanation for the doctor’s consideration, yet she does not expose herself to disaffiliative treatment by compelling the doc- tor’s immediate assessment. The first part of this explanation over- laps the doctor’s immediate pursuit of a candidate location for the discomfort: “In the front.” (line 3). Ms. B then offers an alternate speculative explanation, “Kidney?” (line 4). Dr. A, in overlap with what appears to be the patient’s continuation of her turn, asks again about the location of the pain (line 5), and points to her own right side. Now Ms. B confirms the location (line 6). In this segment, accordingly, the doctor strongly orients to gathering empirical infor- mation before engaging in analysis. (16a) [7:365] 1 Ms. B: .hhh An:::d then I get a lot of tenderness: in this area hh. 2 And again:. It’s probably: (1.0) [whether: it’s] in the = 3 Dr. A: [In the front?] 4 Ms. B: = gall bladder? Kidney? [Er 5 Dr. A: [Up in here. 6 Ms. B: Yeah. Like under the r:ib. Where I can’t get- >it’ll get-< 7 (1.0) very sore. 8 (0.8) 9 Ms. B: ◦.hhh◦ Ptch [A::nd, hhhh ] 10 Dr. A: [‘Bout how often does] that come. 11 Ms. B: Uh:: hhhh (1.0) This cn: (1.5) m- be like at least once or 12 twice a week. And I’ve been trying to see if I’ve been::: 13 >you know,< lifting something or doing something. ◦.hhhh◦ 14 (1.5) ((Dr. A gazes at patient, then turns to record)) 15 Dr. A: How long does it last when you g[et it. ] 16 Ms. B: [Ah::m] (.) maybe a day or 17 two. Furthermore, the doctor maintains her focus on the patient’s expe- rience of the tenderness, asking her how often it occurs (line 10).20 After Ms. B replies (lines 11–12), she reports monitoring her activities for another potentially related event ((lines 12–13); see excerpt 8). There is a silence, where the doctor initially looks at 20 This query (line 10) overlaps with the patient’s line 9 utterance, where she again consults her slip of paper and projects the introduction of a complaint about another (different) symptom. 144 Jeffrey Robinson (line 6) – and she claims she is “not sur:e” in light of the follow- ing evidence: she experienced pain when she “tried (0.2) pressin:g” (lines 6–8). She thus suggests that the evidence points to a more seri- ous problem than hemorrhoids. After a silence (line 9), she begins a repeat of what she “tri:ed,” and during a hesitation in this utterance Dr. D seeks clarification by offering a candidate characterization, “pressing with your hand?” (line 11). Dr. D thus focuses on the evidence that the patient has reported. (17a) [10:594] 1 Ms. N: While we’re on my gut. 2 Dr. D: ◦Yes.◦ 3 Ms. N: A couple a weeks ago: hh u:m (0.6) I had (.) tremendous 4 amount of rectal pai:n? 5 (1.0) 6 Ms. N: No:w- whether it was hemorrhoids or not I’m not sur:e 7 because there was a lot of (0.8) .h pai::n when I tried 8 (0.2) pressin:g. 9 (0.8) 10 Ms. N: tch When I tri:ed hh[a:h 11 Dr. D: [(y’mean) pressing with your hand? 12 Ms. N: No. When I tried ta have a bow:el movement. [(Just-) ] 13 Dr. D: [Pushing] down? = 14 Ms. N: = Yea [h 15 Dr. D: [>Yea< 16 Ms. N: U:m (1.3) Plus there was pain on the outside too, In line 12, Ms. N corrects Dr. D’s candidate characterization by clarifying that “pressing” referred to her efforts to have a bowel movement. In overlap, Dr. D produces another clarification request (line 13). Ms. N confirms this (line 14). Accordingly, while Dr. D exhibits responsiveness to the evidence, this leads them away from assessment and keeps them in the mode of seeking and provid- ing information. In line 16, after Dr. D has acknowledged (line 15) her confirmation of his candidate clarification (line 14), Ms. N returns to describing her symptom, now in terms of where the pain occurred. In subsequent talk (seven lines of transcript not reproduced here), Ms. N describes applying a medication that took some of the “pain and itch” away. She then reintroduces her explanation (line 1 below), expressing a hope that “it wa:s just hemorrhoids.” Explaining illness 145 (17b) [10:613] 1 Ms. N: An I’m hoping it wa:s just hemorrhoids 2 (0.4) 3 Ms. N: >because it< really did ◦hurt a lo:t◦ 4 (0.7) 5 Ms. N: It’s not as bad now. With this expression Ms. N implies its converse: that the prob- lem may be more serious than hemorrhoids. Her next utterance, a reassertion of how painful the condition was (line 3), underscores her implied proposal. She then claims that the symptom has since abated (line 5). Then (not in excerpt here), Ms. N states that she bought a new medication (Anusol), and Dr. D queries further about the symptoms and their location. Reporting that the pain occurred not just externally but also “up some (0.7) in the re:ctum” (lines 1–3 below), Ms. – next proposes (in a qualified manner) that the pain could have been caused by an obstruction (lines 3–4). Again, Dr. D pursues more information about the location of the pain (line 7): (17c) [10:630] 1 Ms. N: >I mean-< there wa:s some external (1.0) pain a:lso but 2 there was a- it felt like it was up some (0.7) in the 3 re:ctum. Um (.) that it was hurting- Almost like it was 4 obstructed there somewhat. 5 (0.5) 6 Ms. N: ◦.hhh ◦ [(N::: ) ] 7 Dr. D: [Could you touch] anything that was tender? Many minutes later, during the physical exam, Dr. D assesses (line 1 below) Ms. N’s original explanation that the pain was caused by hemorrhoids: (17d) [22:1435] 1 Dr. D: (You do have) a fresh hemorrhoid here An[na = 2 Ms. N: [I do, 3 Dr. D: = over on the right si:de. 4 (0.5) 5 Dr. D: (They) also all around the anus it’s very re:d. = h 6 (0.4) 7 Ms. N: Well it has been sore the[re 8 Dr. D: [An I think that’s (probably from 146 Jeffrey Robinson 9 your diarrhea) . . ((11 lines omitted)) . 20 Dr. D: Al:right well we’ll jus’ stop right there Anna cuz I think 21 we know what’s goin’ on 22 Ms. N: What 23 Dr. D: You Have A: hemorrHOID [jis like you] said 24 Ms. N: [Oh ◦okay ◦] 25 Ms. N: I thought you were going to say something to ◦scare◦ me 26 Dr. D: ◦be ◦ a good DOCtor Anna we’ll hire YA 27 Ms. N: ((laughs)) Note that Dr. D constructs this diagnosis as a confirming assessment (“You do have a fresh hemorrhoid”, line 1, and . . . “jis like you said,” line 23), suggesting that the patient’s explanation was correct. In this way, he pitches the diagnosis as responsive to that explanation and also as strongly affiliative. However, in light of Ms. N having designed her explanation in a way that suggested the pain was too severe to reflect hemorrhoids and may represent an obstruction, the doctor’s response is also an oblique disconfirmation of that alternate possibility. Ms. N’s response (line 24) exhibits a “change of state” in her understanding (Heritage 1984b). At line 25, she also displays an “At first I thought (X)’” orientation, implying relief at this dis- confirmation and the realization it entails, a less serious diagnosis (Sacks 1984:419; Halkowski this volume). Conclusion During the information-gathering phase of medical interviews, if the focus typically remains on patients’ symptoms and medically defined ways of exploring them, this is not wholly doctor-induced, nor a matter of doctors imposing a biomedical agenda upon patients who have little agency or understanding of medicine or the medical inter- view. Previous studies have failed to appreciate patients’ dilemma of needing to offer their explanations in a relevant sequential envi- ronment while not disrupting the information-gathering phase of the encounter. Nor has previous research appreciated the doctor’s dilemma of how to receive such explanations before gathering all the data necessary for analysis. Accordingly, the apparent struggle Explaining illness 149 by proposing to examine the breast later in the interview, during the physical exam. She then continues querying the patient about her chest pain; see excerpt (15). (18) [5:271] 1 Ms. A: An’ the only other thing I was thinking of about da:: (.) da 2 pai:ns is if- that could had (.) anything to do: too is with 3 the breast. 4 (2.2) 5 Ms. A: When I had u::m (0.5) ◦pt◦ (0.8) tha:t (0.3) surgery da 6 tumor removed. If dat could be anythin:g (.) connected with 7 that. 8 (1.2) 9 Dr. A: Ptch .hh (.) ah:: >are you hav<ing:: (.) tenderness in in 10 your brea[st it ]self? 11 Ms. A: [Mm hm?] 12 Ms. A: Mm hm = 13 Dr. A: = ◦You are.◦ (.) ◦I’ll check that breast again: an see.◦ Finally, after collecting data in the medical interview and arriv- ing at a diagnosis, doctors could attune themselves better to the patient’s dilemma by following up with a response to the patient’s explanation, such as a confirming or disconfirming assessment, as in excerpts (16b) and (17d). Others have suggested strategies that enhance the patient’s partic- ipation in the interview.22 For example, Lipkin, Frankel, et al. (1995) recommend that the concluding tasks of the medical interview be reconceptualized so that, besides doctors delivering information to patients, patients themselves enter the analytic discussion with their perspectives and beliefs. Physicians should ask patients what they have understood about the physician has told them. Lipkin, Frankel, et al. (1995:79–80) write: In the course of the patient’s exposition of what he or she has understood, the patient will reveal his/her explanatory model of the illness process – that is, to what the patient attributes the problem. These so-called attribu- tions, the patient’s sense of the meaning or causality of the illness, must be acknowledged or some patients will reject the clinician’s approach as not appropriate. 22 See the discussion of cultural influences on patients’ “explanatory models” in Johnson et al. (1995). 150 Jeffrey Robinson We agree that physicians should encourage patients’ participation in medical interviews. Our analysis, however, shows that patients may reveal their explanatory theories well before the closing moments of these interviews. Patients orient to the process by which data is gath- ered and analyzed; in overt and tacit ways, they offer their explana- tions in information-gathering locations such that doctors can con- sider them as they generate diagnostic hypotheses. To suggest, as Lipkin, Frankel, et al. (1995) do, that doctors should elicit patients’ participation in the analytic discussion at the end of the visit, is to assume that patients will believe that doctors can take their theo- ries into account as possible candidate diagnoses at this point, even though they have finished collecting data and have already delivered a diagnosis. It is perhaps more likely that patients may interpret the doctor’s efforts to give them (what Lipkin, Frankel, et al. [1995] call) a “final shot” at determining the agenda of the visit, as a move designed to make the doctor appear responsive. In addition, with this strategy doctors miss the opportunity to take patients’ explana- tions into account while considering and testing various diagnostic possibilities. We therefore suggest that if a patient does not volunteer a causal explanation within the data-collection phase of the medical interview, the doctor should solicit the patient’s explanation in that location, rather than wait until the end of the visit.