A direct observation of pain scale use in five video-recorded palliative care consultations: Using conversation analysis to show how practitioners support patients to describe pain.

OBJECTIVE: Assessing pain intensity is an important palliative care task. Self-report pain intensity scales are frequently used within assessment. In contrast to formal studies of validity and reliability, we examine administration of, and responses to these scales in everyday palliative care.
METHODS: We searched episodes of pain scale use in a dataset of (video/audio-recorded) UK palliative care consultations involving five doctors, 37 terminally ill patients and their companions. We found five, and applied the techniques and tools of conversation analysis to characterise scales' administration and functioning.
RESULTS: Generally, the patients responded to scales by reporting multiple aspects of pain; the doctors supported and encouraged this. In two episodes, the scales generated misunderstandings. The doctors worked to resolve these in ways that avoided implying the patient was at fault.
CONCLUSION: Pain intensity scales can yield richer information than just intensity. They can also generate misunderstandings and social friction which take skill and effort to resolve. PRACTICE IMPLICATIONS: Patients tend to respond to pain intensity scales by reporting on multiple aspects of pain, professionals should support them in this. These scales sometimes generate misunderstandings. To preserve the therapeutic relationship, professionals should work to resolve these without implying the patient is to blame.

Introduction

Patients with life-limiting illnesses report that pain is one of their most distressing symptoms [1]. Assessing pain in palliative care is central to effectively addressing symptoms and monitoring illness progression, as signalled by deterioration or increase in symptoms [1], [2], [3]. Pain intensity is argued to be the most clinically relevant dimension of the pain experience [4].

Literature and clinical guidance converge in advocating for standardised, unidimensional tools to measure self-reported pain intensity with numerical, visual analogue, and categorical (mild, moderate, severe) rating scales [4], [5]. These measures perform well in formal tests of usability, reliability, and validity for the purposes of research and experimental settings [3], [4]. However, to date, no studies have reported on whether in everyday clinical practice, pain intensity scales work in the way they are supposed or assumed to. Use of pain scales warrants further examination for two reasons. Firstly, studies of other standardised assessments show that formal guidance and actual practice diverge in ways that impact the way patients respond [6], [7]. Secondly, patients with advanced disease describe pain’s complex impact on performing everyday tasks and maintaining relationships with loved ones [8], [9]. Patients report difficulties responding to pain intensity scales [8], but there have been no studies directly examining this. Given all this, we ask:

What opportunities and obstacles arise when administering pain intensity rating scales in real life palliative care consultations?

Methods

The data come from 37 consultations video/audio recorded in 2014, involving five experienced palliative medicine doctors and 37 patients (in 16/37 companions also attended) who provided written consent. Consultations took place at a large U.K. hospice providing care and support to terminally ill patients, their family, and carers. Information on participants and data collection has been published elsewhere [10]. We identified episodes in the recordings and verbatim transcripts wherein pain was mentioned (see Jenkins et al. [11] for an analysis of pain question sequences in this dataset), and wherein doctors asked patients to rate their pain using a scale. These episodes were subjected to conversation analysis: a rigorous approach to naturally-occurring talk which builds empirically-grounded descriptions and understandings about how doctors and patients interact [12]. Analysis included Jeffersonian transcription [13] of all pain scale episodes, capturing fine details such as intonation, emphasis and silences.

Results

Most consultations (34/37) included references to pain, overwhelmingly initiated by doctors (n = 180). Pain was also introduced by patients (n = 40), largely following doctors’ non-pain specific turns, such as problem presentation solicits and enquiries about general symptoms, functioning, or medication. Occasionally, pain talk was raised by companions (n = 12) (Table 1 ).

Table 1

Distribution of speakers instigating pain talk across the dataset.

Pain talk	Initiating position	Responsive position	Total
Doctor	180	0	180
Patient	2	38	40
Companion	4	8	12

In five consultations, doctors administered numerical and/or lexical pain intensity scales. We report our two key findings: firstly, that patients communicate not only their pain’s intensity but other aspects too - and that doctors deployed communication practices which supported them to do so; secondly, misunderstandings about numerical scales occurred (2/5 of our episodes) and were handled with sensitivity by the doctors to avoid laying blame for the misunderstanding at the patient’s door. We present extracts that represent patterns found across our small collection.

Communicating more than pain intensity

In our first example, a male patient with advanced cancer is attending the outpatient clinic with his brother. When discussing the patient’s lower back pain, the doctor introduces a numerical scale.

The doctor asks three linked questions (lines 162, 175–6, 180) about: current pain; worst pain; and pain at its best. Elements of a question specify what the recipient should do in their response [14], [15]. Each question’s design – its grammar and wording – (“how bad…if you sometimes score it…”, “how much is it”) clearly conveys that the expected, most relevant answer is a number. The doctor seems to be heading towards an explanation of ten being the most pain, but he stops explaining when the patient begins to answer (line 164), showing familiarity with the scoring system. As a result, the pain scale is not delivered in a standardised manner: to do so the doctor would have to talk at the same time as the patient. Instead, we see that the delivery of the pain scale is sensitive and calibrated to the specific interactional context; the doctor allows the patient to deliver his answer without obstruction.

Despite the pain scale inviting single number responses, the patient provides fuller answers to each: “At the moment, now, maybe three four”; his worst pain is “Seven eight nine.” (line 177), and best: “two three.” (line 182). He thereby conveys the intensity of his pain as variable, not static. We see this in our other episodes, for instance in one consultation, the doctor asks about whether the pain was “mild, moderate or severe” and the patient’s answer: “moderate at times” conveys fluctuating intensity.

After the patient has responded to the first pain question, the doctor’s brief ‘yeah’ and ‘okay’ responses and the short silence he lets develop (line 168) provide the patient with opportunities to say more. The patient continues (lines 173–174): “Because I’m sitting down und… talking to you.” and “Because I’m feeling out of breath as well but that’s due to my lung definite.” In this, he connects pain intensity with aspects of functional capacity (his physical position and breathing), aspects that patients often report to be an important part of their pain experience [8]. This clinically useful information builds a broader depiction of his pain.

In this and other episodes we collected, the patient capitalises on the way that introducing a scale opens up opportunities to expand, and the doctor provides interactional space for these expanded responses.

Misunderstandings about numerical scales and their careful handling

The doctors in our recordings work towards joint understanding about the pain scale. Nevertheless, misunderstandings about numerical scales occur (2/5 of our episodes). Our second example illustrates this.

The doctor has already asked the patient (who has progressive and advanced cancer) “how difficult is the pain if you were scoring out of ten in the morning … If ten’s the worst.”. The patient rated her current pain as seven. Atypically for our recordings, she provided one single number, but like Extract 1 and others in our data, she qualifies her response with additional information: “it’s ea:sed it’s: you know it’s really (0.4) helping”. In this way she similarly builds a broader picture of her pain experience. We join the consultation a few minutes later as the doctor asks about the pain level at its best.

Having previously rated her current pain as seven, the patient now scores her best pain as eight and a half, revealing that she understands scores nearer to ten to mean less pain. The doctor’s “Say that again sorry?” (line 60) shows the patient that there is some kind of problem [16]. In response, she firstly repeats and slightly modifies her score, treating the doctor as not having fully heard. She then qualifies her response “I mean it is a lot better”, “possibly nine” (lines 64–66).

At this point (line 67), the doctor begins to point directly to the misunderstanding: “I think we’ve got scales” but stops as the patient continues to describe her pain. By doing this, he enables her to continue providing clinically relevant information, and this leads on to a discussion about her pain medication. About a minute later, we re-join the consultation in extract three as the doctor returns to the misunderstanding.

The doctor points out their divergent understandings of the pain scale. Importantly, he works up a claim that he is to blame for the misunderstanding – and thus that the patient is not: “I think I’ve got- I’m at cross purposes. I think I’ve confused things.” (lines 89–90). The doctor’s work to avoid laying responsibility for misunderstanding on the patient is all the more evident when her husband counterclaims that the patient was “going the wrong way” (line 98), which the patient acknowledges on line 100. After this (data not shown), there are several unsuccessful attempts to reach joint understanding and get numerical responses from the patient; eventually, the doctor introduces a mild/moderate/severe scale and this meets with success.

In our data, numerical scales can engender misunderstandings which risk inaccurate assessment. Pointing out misunderstandings also carries risks of making the patient feel stupid and damaging the therapeutic relationship. We have shown that it takes time, as well as interactional effort and skill to resolve.

Discussion and conclusion

Discussion

Unidimensional scales of pain intensity are designed to elicit patients’ pain in a standardised and reliable manner. Our analysis of five video-recorded examples of actual pain scale use in palliative care produced three key findings, available to be explored with broader datasets across a range of clinical settings.

Firstly, as with other standardised assessment tools [6], [7], practitioners may diverge from standardised delivery of the pain scale. We demonstrate good clinical justification for this: if patients demonstrate understanding and begin to answer, continuing to explain the pain scale would come across as confrontational or patronising.

Secondly, although pain intensity scales are designed to elicit a single number or word, patients frequently respond with two or three numbers or additional descriptors of the pain experience; professionals can support these clinically useful responses which have been identified as important to patients [8].

Thirdly, patients report difficulties responding to pain scales [8], [9]. Misunderstandings occurred in our data. They carry risks of inaccurate clinical assessment and interactional awkwardness and embarrassment. Resolving those misunderstandings whilst circumventing those risks takes time, communicative skill, and effort.1

Conclusion

Directly observing what happens when standardised pain intensity scales are administered in real, everyday clinical interactions, shows that they are administered in ways that diverge from the recommended administration, and that there are good clinical reasons for doing so.

Practice implications

Practitioners can show sensitivity to how the patient is responding, and carefully handle misunderstandings in ways that support and protect the clinical relationship and provide for more comprehensive descriptions of the pain experience.

Ethical consent

I confirm all patient/personal identifiers have been removed or disguised so the patient/person(s) described are not identifiable and cannot be identified through the details of the story.

Ethical approval was obtained from NRES Committee West Midlands—Coventry & Warwickshire, UK (ref 14/WM/0128).

Funding sources

The authors disclosed receipt of the following financial support for the research, authorship and publication of this article: The preparatory stages of this work were supported by the University of Nottingham, including via a Research Development Fund grant from the University of Nottingham’s Centre for Advanced Studies; later stages of work were funded by The Health Foundation [Grant ID RU33/GIFTS 7210], an independent charity committed to bringing about better health and social care for people in the UK. From October 2015 onwards, Parry’s work, including in preparing this report, has been supported by National Institute for Health Research Career Development Fellowship Award [Grant ID CDF-2014-07-046]. This report presents independent research partially funded by the (NIHR). The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health.

CRediT authorship contribution statement

Laura Jenkins: Conceptualization, Data curation, Formal analysis, Funding acquisition, Methodology, Project administration, Writing – original draft, Writing – review & editing. Ruth Parry: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Writing – review & editing. Christina Faull: Data curation, Funding acquisition, Investigation, Project administration, Resources, Writing – review & editing.

Declaration of interest

The authors declare no conflict of interests.