Social Representation of Pain and Suffering in Cancer Patients: A Mixed Methods Study

Abstract

Most patients report physical pain and psychosocial problems (suffering, depression, anxiety) during and after cancer treatment. This mixed methods study, based on the Theory of Social Representations, examines if pain and suffering are taboo subjects for oncologists and patients, and if they share the same representations. The data collected included 33 in-depth interviews with patients in chemotherapy/radiotherapy treatment (n = 20) and their oncologists (n = 13). We measured the number of spontaneous mentions of the words “pain” and “suffering” and their synonyms. We then examined semantic networks related to these two terms. The results show that for patients, suffering refers to bodily degradation and vulnerability while pain confirms the presence of the disease. In the interviews conducted with oncologists, suffering is absent. Pain is perceived as a sign of disease progression and an indicator of tolerance to treatment. These results may help in increasing the mutual understanding between oncologists and patients and facilitating the detection of depression and anxiety.

Keywords

pain suffering discourse analysis mixed methods Social Representations

Introduction

Pain and Suffering in Cancer Management

Cancer pain management is a major concern for professionals (Brennan et al., 2007; Nakash et al., 2012) and is well documented in the literature (Makhlouf et al., 2020). However, despite the increasing attention given to pain management, pain remains one of the most commonly encountered physical symptoms in cancer care. According to a recent meta-analysis, more than 50% of patients report experiencing pain (Den Beuken-van Everdingen et al., 2016). Rodriguez and Van Cott (2005) underline that healthcare providers and patients must communicate about pain, especially as it cannot be directly observed.

As noted by Granek et al. (2020), the treatment of mental health distress in cancer care has several positive outcomes, including, for example, in the progression of the disease, the length of hospital stay, treatment adherence, satisfaction with treatment, and quality of life. However, suffering is underdiagnosed in cancer treatment: 30% of patients develop an anxiety or mood disorder (Nakash et al., 2013). For example, in the case of colorectal cancer, Rozmovits and Ziebland (2004) showed that patients refer to loss of self-respect, privacy, independence, and sexual self-assurance, as well as their work and their relations with other people. Philippini (2009) noted that it is not easy to diagnose a patient’s suffering, and even less a state of anxiety or depression, because this requires questioning the boundary between a normal and a pathological reaction. It is therefore possible that medical professionals do not pay enough attention to the suffering of patients.

In the literature, a distinction is made between pain and suffering. This distinction refers to a classic medical conception that people are divided into two entities: body and mind (Bosire, Mendenhall, & Weaver, 2020). Pain refers to physical or somatic aspects, whereas suffering is a psychological perception related to the experience of pain (Bouckenaere, 2007). According to Franck (2001), suffering is radically different from pain because it refers to the feeling of being disconnected from one’s previous life and becoming a different person. Suffering is the feeling of a loss of what could have been right if something wrong had not happened. More recently, in the human sciences, the distinction made between pain and suffering is less radical. For example, the concept of “global pain” defined by Clark (1999), Krawczyk et al. (2018), and Pinto et al. (2019) removes the boundaries between pain and suffering. For these authors, pain is more than a physical matter of sensation since it includes the interrelations between social, physical, psychological, and spiritual dimensions. The concept thus allows for the person as a whole to be taken into account and provides a means of overcoming a hyper-specialized care system that tends to focus on the disease rather than the patient.

These different conceptions of pain and suffering led us to question the representations that patients and oncologists have of these two objects. Do they share the same representations? Are there gray areas or taboos in these representations? Do oncologists have a clear perception of what patient suffering means? A weakly structured representation of pain and suffering may be one of the reasons for the under diagnosis of mental health distress in cancer care.

Theory of Social Representations

A mental representation designates both the action of representing and the product of this action (Clapin, 2002). In the latter sense, a representation refers to a mental image constructed to replace (i.e., to re-present) an object when this object is absent. The image (the representing world) differs from the object (the represented world). It is therefore a subjective reconstruction depending on the object’s characteristics, the subject’s knowledge, and the context (Hubbard, 2007).

The model of mental representations does not take into account the group influence on this reconstruction. On the contrary, Social Representations (SRs) concern shared cognitions that impact individual behaviors and decisions (Abreu-Lopez & Gaskell, 2015; Moscovici, 1984, 2000). According to Moscovici (1984), SRs acquire meaning through social group relationships and verbal interactions. Professional Representations (PRs) are part of an SR. They include academic knowledge constructed during professional training (Morant, 2006).

The structural approach of SRs advocates that cognitions are split into two systems. Cognitions in the central system are strongly consensual; they refer to ideas, values, and practices that are fundamental to the group since they are linked to norms and values. Conversely, cognitions in the peripheral system are more individual. They make concrete the central cognitions in a person’s daily life. Central cognitions structure the SR. In other words, they are strongly connected with each other and with peripheral cognitions. When a central cognition is evoked, so are the others. Central cognitions structure the semantic network within discourse. They can be represented as nodes in a Maximum Tree (MT), that is, a star-structured network of cognitions strongly correlated with each other (Abric, 2005).

Professional Representations (PRs) of an object can be radically different from SRs because of the specialized knowledge gained by the professional during training. Comparatively, SRs contain common sense knowledge. For example, Jeoffrion et al. (2016) crossed the representation of illness among health professionals (HPs) and non-health professionals (NHPs) and confirmed this result. Their analysis also showed that the HP’s representation of illness is more highly structured than that of NHP. The central cognitions are strongly shared and related to the diagnosis and consequences of the disease. The peripheral cognitions refer to the illness’s characteristics.

Researchers have several methodological options when collecting SRs. Qualitative methods reveal the semantic networks in discourse and thus the participants’ underlying logical thoughts.

Pain and Suffering as Objects of SRs

A study by Dany et al. (2006), conducted through interviews with 25 patients and 30 female nurses, examined the SR of suffering. The authors used an Interpretative Phenomenological Analysis. The results showed that neither group drew a line between pain and suffering. The authors noted that suffering encompasses pain but is broader and includes a social component, implying repercussions on the immediate social environment (i.e., a change of status within the family and at work) as well as on other social relationships (other patients, health professionals, etc.). Finally, suffering was also found to have an existential, spiritual, and psychological dimension reflecting a sense of psychological malaise and raising questions about the role of humans in the life cycle and about religion.

Péoc’h (2014) showed that patients experiencing chronic pain have SRs of pain structured around four thematic classes: (1) Existential pain refers to the loneliness experienced by patients in pain, causing them to isolate themselves and to feel isolated; (2) pain state refers to physical and bodily injury or damage and is the sign or manifestation of that injury or damage; (3) burdensome pain refers to a situation of impotence experienced by the patient, who is lacking, and incapable of regaining, control and feels overwhelmed; and finally, (4) shared pain refers to the shared nature of pain that affects us all when we experience illness.

According to Fromage and Hatti (2015), pain and suffering are clearly intertwined. The authors argue that while a painful experience is embedded in a physiological substrate, its expression is the result of an interaction with the psychological characteristics of the individual assigning meaning to it. Thus, in some cases, psychological pain can transfer to the bodily realm. For example, in cases of physical scarring, subjects often experience psychological tension, which they relocate by transferring it to the body, thereby enabling them to regain control. Conversely, in palliative care, a patient’s pain can be amplified by anxiety, and the expression of pain may serve to justify a request to be listened to.

These studies are of great value in demonstrating that in patients’ SR, pain is not associated with a physical phenomenon and suffering is not independent of the pain felt and experienced by individuals. However, none of these studies seek to compare patient/oncologist representations with a view to identifying areas where they converge and diverge as well as the semantic network structuring each representation. This is, however, of great interest considering the medical care of patients.

Developing the idea that there is a deficiency in patients’ psychological care in oncology, we decided to measure how oncologists perceive pain and suffering compared to patients. Our approach was guided by two questions: Are pain and suffering taboo subjects, and so, do oncologists and patients evoke spontaneously these objects? Do they have the same representations of them?

To maintain objectivity and avoid researcher biases, among the various analytical methods employed for analyzing SRs, we used content analysis software and a Similarity Analysis (SA) method. SA is based on a correlation analysis that quantifies the number of connections between each word embedded in a square matrix. This correlation analysis enables the words that are central in the semantic network to be distinguished from those that are peripheral. The central words have the greatest number of links and the strongest correlations with the other words (Abric, 2005).

Our objectives were thus threefold: (1) to compare the SR of cancer among patients and oncologists; (2) to evaluate the place occupied by pain and suffering within patient and oncologist representations; and (3) to reveal the semantic networks related to pain and suffering in the form of a Maximum Tree (MT). Finally, we sought to examine the implications for the mutual understanding between oncologists and patients as well as the consequences on overall patient care.

Method

Study Design and Participants

This study was part of a larger project named EOLE, which is looking at how a therapeutic decision is co-constructed between the oncologist and their patient. It was approved by the Research Ethics Board prior to launching the study (IRB approval in September 2017: AU 1338/2017-A00776-47¹).

Thirty-three participants took part in this study: 20 patients and 13 oncologists. All the participants were interviewed on six sites specializing in oncology in France.

Procedure

At the first stage of the procedure, 20 oncologists were contacted. They were informed of the background and objectives of the study, and only volunteers were included. Seven of them declined to participate, stating a lack of time.

At the second stage of the procedure, 31 patients of these oncologists were contacted after a medical follow-up visit. It should be noted that during their treatment, all the patients attended their oncologist every week, over a period of 6–12 weeks. During this first contact, patients were informed by a technician from the hospital about the aims of the study and the guarantees of anonymity. The professional in question was experienced and was responsible for recruiting patients for the studies carried out in various departments.

Patients were also told that their answers would have no impact on their treatment. Eleven of the 31 patients declined to participate, stating fatigue and a lack of time. When the patients gave their consent, the interview took place either immediately or no later than the next consultation, within 15 days. All the interviews with oncologists and patients took place in hospital. They were undertaken by graduate students in social psychology with experience of conducting interviews in health research contexts. The technician responsible for obtaining consent and the interviewer were thus two different individuals, thereby ensuring that the offer to participate in the study was not experienced by the patients as a pressure.

Interviewer Training

The study principal investigator trained the four interviewers, who were graduate students in social psychology with experience in carrying out interviews in health research contexts. Interviewer training lasted 3 days and included interviews with volunteer patients and oncologists. At each training session, feedback and analysis of the interview technique were provided.

Course of Interviews

The participants took part in one-hour, semi-structured individual interviews. The interview guide is reported in Table 1. Each interview began with opening remarks designed to set out the communication contract between the interviewer and the interviewee. The interviewer was introduced, and the subject, framework, and objectives of the study were then specified, as was the length of the interview. To standardize the interviews and limit information biases, the introduction was written and memorized by the interviewer.

Table 1.

Participant Demographics: Oncologists (n = 13) and Patients (n = 20).

Characteristics	Oncologists (n = 13)	Patients (n = 20)
Years of practice, M (SD)	15.4 (5.6)	x
Gender, % (n)
Male	46 (6)	60 (12)
Female	54 (7)	40 (8)
Age, M (SD)	40.8 (5.6)	67.2 (9.4)

Consistent with ethical guidelines, the participants were asked to give their explicit consent to participate in the study and reminded that they were entirely free to agree or refuse to take part. The interviewer specified that declining to take part would not impact the interviewee negatively. All the interviews went well. However, in the event of a patient feeling unwell, we planned to refer them to the psychologist working at the hospital center with which we were in contact.

To ensure that the focus of the interviews was on Social Representations, the participants were subjected to a social identity anchorage in relation to their social group: the oncologists were interviewed “as oncologists” and the patients “as patients.” The opening question was deliberately broad and concerned the representation of cancer. This first part of the interview was unstructured. The interviewer repeated and summarized what the participant said and formulated questions seeking clarifications.

General questions presenting new themes (cf. Annex 1) were included in an eight-section interview guide. They focused on (1) the history of the disease; (2) the description of disease management; (3) the representation of therapeutic treatment; (4) the oncologist’s role in therapeutic decisions; (5) the patient’s role; (6) the relative’s role; (7) a Word Association Task: uttering the first five words that came to mind when thinking about therapeutic decisions; and (8) the interviewer asked participants if they had something to add.

Data Analysis

The interviews were audio-recorded and analyzed using IRaMuTeQ® 07 alpha (2020). This is content analysis software used to highlight the structure of a discourse through correlational statistical analysis represented visually in a graph known as a Maximum Tree (MT).

(1) The corpus of the text was first disambiguated: all personal pronouns, possessive or demonstrative adjectives referring to the target referents (oncologists, patients), were replaced with the appropriate referent; (2) the sub-corpus related to the representation of cancer was analyzed; and (3) two sub-corpuses from all the interviews, one of which related to “pain” and the other to “suffering,” were identified and analyzed.

To restrict the corpus to full words only, an analytical key was applied and only nouns, adjectives, and verbs were retained. Finally, a level-one lemmatization process was performed on the sub-corpus. Words sharing the same base form were grouped together (e.g., pain and painful; suffering and suffer) and an SA was then performed. A correlation matrix was calculated. All the words in the corpus were identified, and the number of connections between them (i.e., co-occurrences) was quantified. The connections are represented visually in the Maximum Tree. The connections between two words are called crests, while words are termed summits. The element with the highest number of crests and the strongest connections is the most central element. They form nodes in the center of a semantic network containing peripheral elements. Quotations of illustrative or typical interview extracts exemplify this semantic organization.

Results

Participant Demographics

Information concerning participant demographics is presented in Table 1.

We looked for maximum variation in the sample by targeting oncologists and patients in six study sites. We were able to balance the gender distribution of the oncologists (n = 13; 6 males and 7 females). They were at different stages of their career (from 5 to 21 years of practice). They were more than 30 years old (aged 31–48; mean age M = 40.8 years). They treated three cancer sites (lung, colon, and breast) and had been practicing for an average of M = 15 years.

We sought to balance the gender distribution of the patients (n = 20). However, lung cancer occurs mostly in males and breast cancer in females. Colorectal cancer is found in both males and females. Thus, 12 males and 6 females were included. They were aged from 58 to 82 years, with an average age of M = 67.2 years.

Representation of Cancer: A Qualitative Analysis

A Similarity Analysis (SA) was performed on the unstructured part of the interviews. Figure 1 presents the MT of the patient interviews, while Figure 2 shows the MT of the oncologist interviews.

Figure 1.

Maximum Tree based on the sub-corpus from the unstructured part of the interviews conducted with the patients (n = 20)—representation of cancer.

Figure 2.

Maximum Tree based on the sub-corpus from the unstructured part of the interviews conducted with the oncologists (n = 13)—representation of cancer.

Figure 1 shows that patients’ SRs of cancer are centered around the central element, doctor, which organizes a semantic network relating to relationships established with the doctor in relation to the treatment (e.g., decision and trust). Five main nodes were identified: (1) the node patient, in direct line with doctor, organizes a semantic network corresponding to the symptoms of the disease (e.g., pain and shaking); (2) the node cancer refers to the social consequences of the disease (e.g., fear, difference, and problem); (3) the node thing refers to the psychological effects of the illness (e.g., depression and shock); (4) the node treatment structures the semantic network concerning practical aspects, including the economic consequences of care (e.g., pill and cost); and (5) the node chemotherapy combines elements related to the side effects experienced from the treatment (e.g., nausea and vomiting).

This MT indicates that pain and suffering are not associated with the same semantic field. For patients, pain is related to the symptoms of the disease, while suffering is related to the social consequences of the disease. These results also underline that the social and psychological consequences of the disease (including work activity) play an important part for the patients. Surprisingly, side effects are the more peripheral elements in the representation of the disease.

As shown in Figure 2, the oncologists’ representation has a different structure and content.

Figure 2 shows that, compared to patients, oncologists have a more structured representation of cancer, as evidenced by the narrowing of the MT around a more limited number of elements and the use of a more specialized vocabulary. Although we observe a large number of common nodes, the semantic networks do not contain the same elements and do not have the same significance.

The oncologists’ PRs are structured around the central element, patient, which is directly connected to elements relating to the therapeutic monitoring of patients. In this semantic network, the elements of the diagnosis announcement, care plan, and chemotherapy are highly connected. From this central element, four nodes emerge: (1) the node treatment structures a semantic network referring to treatment methods and effects (e.g., toxicity and aggressiveness) as well as the adaptations needed to respond to these effects (e.g., resistance of the tumor and adjustment of the treatment). Unlike with the patients, “chemotherapy” was not found to be a stand-alone node; (2) the node illness combines terms referring to the physical effects of the treatment (e.g., fatigue). Unlike with the patients, psychological consequences are not cited; (3) the node cancer articulates a semantic network related to the characteristics of the disease, its duration, and location (e.g., lung); (4) the node doctor articulates terms referring to the role and position of the oncologist in the relationship with the patient (e.g., case and response) and, more particularly, patient management (e.g., argument and personality).

Based on Figures 1 and 2, we may conclude that patient and oncologist representations of cancer share four common nodes (patient, doctor, cancer, and treatment) and three different nodes (disease, chemotherapy, and thing). However, significant differences appear.

Patient SRs were found to be structured around the central element doctor, while oncologist PRs were structured around the patient. In other words, each group tends to represent illness and disease in terms of the other group, with oncologists referring to the patient’s disease and patients emphasizing the disease treated by their doctor and which they are experiencing.

The patients tend to refer to the social and psychological consequences of illness, whereas the oncologists place greater emphasis on the physical consequences.

Finally, with respect to our research question, pain and suffering were not found in the oncologists’ representations. The two concepts were found to correspond to different semantic fields among the patients, with suffering being connected to cancer and, more specifically, to the social consequences of illness, while pain was connected to the patient and, more specifically, to illness symptoms. We found the same duality referred to above between the psyche and the body, as illustrated by the following quotation from a patient’s interview describing the interrelations between pain and suffering.

It [cancer] changes a lot of things (…) you have a social life, (…) and it suddenly stops (…). It’s complicated (..) it all sort of piles up. There’s the physiological side of things, pain, difficulties … And then there’s the social side (…). I’ve lost friends because it scares them, and because I wasn’t around to see them at the time because I was suffering. (Patient 8)

In short, in patients’ SRs of cancer, pain is related to suffering because constant discomfort creates a gap between the patients and their relatives. The patients are isolated because they focus on their pain and because relatives distance themselves as they fear illness.

Place of Pain and Suffering in the SR of Cancer: A Quantitative Analysis

We aimed to complete the qualitative analysis to verify if pain and suffering were taboo words in the discourse of patients and oncologists. We focused on the occurrence frequency of pain and suffering across the entire corpus. The average number of occurrences of pain (or the corresponding adjective, “painful”) and suffering (or the corresponding verb, “to suffer”) was counted. A single-factor analysis of variance (ANOVA) was then performed to compare the two groups of participants (Figure 3).

Figure 3.

Comparison of the average number of references to pain and suffering in the interviews conducted with oncologists (n = 13) and patients (n = 20).

The ANOVA revealed a significant difference in the average occurrence frequency of the word “suffering.” The patients used the term more frequently (M = .30; SD = .70) than the oncologists (M = .23; SD = .87) F(1,31) = 1.36, p < .02, η2 = .16. By contrast, no significant difference was found in the use of the term “pain,” with the oncologists (M = 3.15; SD = .84) using it just as often as the patients (M = 2.15; SD = .68).

In other words, while the patients referred clearly and explicitly to suffering, that is, the psychological experience of pain, there was little mention of either among the oncologists. To complete this analysis, we examined the semantic networks linked to these two elements.

Semantic Networks Associated with Pain and Suffering: A Qualitative Analysis

To understand the meanings associated with pain and suffering by both groups of participants, a sub-corpus containing these targeted words was extracted and an SA was then performed. The SA is presented in Figure 4(a) and (b) and Figure 5.

Figure 4.

Maximum Tree based on the sub-corpus from the interviews conducted with the patients (n = 20): (a) on pain (Fc = 47) and (b) on suffering (Fc = 62).

Figure 5.

Maximum Tree—doctor interviews (n = 13) on pain (Fc = 51).

Patient SR of Pain and Suffering

The MT presented in Figure 4(a) is related to pain. It shows an SR without nodes. Three elements are apparent: area; doctor; and go. A verbatim quote will serve to illustrate their semantic organization. The words that appear in the MT are underlined.

I’d (patient) sometimes have stomach pains (…). Then I (patient) got little aches and pains in the testicle area. (…). I (patient) told him (doctor) I (patient) sometimes get pains in my stomach. He (doctor) gave me (patient) a scan. (…). They’d found cysts everywhere, in the groin area, the stomach area, and in the neck area. And then [sigh] I was given a treatment that also caused unpleasant side effects (…). It’s not really a question of being a bother though, it’s more a matter of pain. For example, with the antibiotics, when you (patient) get diarrhea or stomach cramps every day, it gets difficult and painful. (Patient 4)

In patients’ SRs of pain, pain is connected to the location of the cancer, making the existence of the disease and its diagnosis a reality. Pain is also related to the unpleasant side effects of the treatment. Therefore, for patients, the pain experienced is caused by both the disease and the treatments.

The MT in Figure 4(b) is related to suffering. It shows an SR without nodes. Five elements co-occur more frequently: patient knowledge; doctor; do; can. The following interview extract illustrates one possible semantic organization. The words that appear in the MT are underlined.

They (doctor) have exhausted all therapeutic options and therapeutic treatments… And there’s nothing else they (doctor) can do … In any case, the illness gets worse, so you (patient) get worse, and I (patient) think I’ve got to a point where it’s all just suffering (…). I’m (patient) not in favor of futile medical care, when you know that ultimately, there’s no way out (…) when they (doctor) carry on with treatments after treatments, while all the time knowing that there’s nothing else that can be done…. So when you (patient) can no longer stand up, when you (patient) can no longer walk, when you’ve (patient) stopped eating, you’re (patient) fed through a, feeding tubes, or drips, I (patient) don’t see the point. Because I (patient) think we (patient) still have our (patient) moments of lucidity, and it’s awful to see yourself (patient) like that… Awful to know also that you’re (patient) just waiting to die. (Patient 18)

Based on the above quote, we may say that for patients, suffering amounts to psychological pain. It refers to a sense of physical deterioration, a poorer self-image, a loss of self-respect, and a feeling of vulnerability to death once the various therapeutic treatments used by doctors have failed. The patients feel alone and isolated; they live with suffering, they are just suffering. Suffering creates a barrier between them and their relatives. To illustrate this representation, we may consider, for example, the following quote from the interview conducted with Patient 4:

What made me cry was a mixture of physical pain, feeling as low as you can get and feeling like I couldn’t have a social life anymore or communicate with people (…). I couldn’t communicate with my wife anymore. And it made me, it destroyed me even more. Because I couldn’t communicate anymore, I didn’t feel like talking anymore, or doing anything. It was total suffering.

Oncologist PR of Pain

Since the term “suffering” did not occur sufficiently often among the group of oncologists, an SA was not performed on this sub-corpus but only on the “pain” sub-corpus. The results are shown in Figure 5.

The MT presented in Figure 5 shows that the representation includes just one node: patient. The other elements are not structured and are directly connected to pain.

Two zones can be identified: (1) treatment and side effects and (2) symptoms and illness. A re-examination of the corpus will help to illustrate this interpretation. Consider, for example, the three following quotations:

We (doctor) wouldn’t want to get them (patient) to undergo radiotherapy just because we’ve found a little image but there isn’t any particular pain or any risks. (Oncologist 1)

There’s the patient’s general condition, their weight, appetite and symptoms such as pain, the distended stomach when acid reappears. The clinical examination showing that the liver is enlarged again although we (doctor) couldn’t feel the metastases anymore. The objective clinical signs that tell us (doctor) there’s something going on. (Oncologist 6)

Because, in some cases, we (doctor) achieve prolonged survivals, so it’s rare that patients don’t want a prolonged survival, whereas short survivals at the cost of side effects and intensive treatments may not be the right thing for everyone. There are some patients who’ll say ‘I don’t see the point of surviving for another 2 months and having to come in every fortnight for chemotherapy, and having nausea, vomiting, and pains’. Prolonging survival is a worthwhile aim but it’s not necessarily the only aim of anticancer treatment. (Oncologist 7)

The above quotes suggest that for oncologists, pain is largely interpreted as one of the side effects of treatments, as a symptom of the evolution of the illness, or in relation to the patients’ response to treatments, that is, their tolerance. Therefore, pain is evaluated by the oncologist according to the benefits of the treatment (i.e., the fight against the disease to make it chronic) and the cost (i.e., deterioration of patients’ quality of life due to the side effects).

Discussion

Pain and suffering lie at the heart of therapeutic care and, more specifically, of therapeutic decisions. However, the representations that patients and oncologists have of these two objects do not overlap.

Patients’ representations are inherently social and constructed through the experience of illness. Oncologists have a professional representation, acquired through their academic training and medical practice. Consequently, in accordance with SR Theory, there is a difference in the content and in the structure of the representations that oncologists and patients have of cancer, pain, and suffering.

Oncologists’ representations of cancer are more structured in the sense of having a narrower focus around a smaller number of elements as compared to patients’ representations. In other words, oncologist representations contain fewer words than those of patients. Moreover, the associated vocabulary is more technical and includes specialized terminology (e.g., toxicity, diagnosis announcement, and care plan). Patients’ representations contain comparatively more words and a vocabulary highly related to the practical consequences of the disease (psychological, physical, social, and economic). These results are consistent with those of Jeoffrion et al. (2016).

Moreover, the illness representations of oncologists and patients do not have the same content. For oncologists, illness is related to the medical treatments, that is, the protocols (method) and their effects. In addition, oncologists mention the physical effects of the disease (e.g., fatigue). Thus, for oncologists, adjustments are necessary in order to adapt the treatments (1) to the aggressiveness of the treatments (toxicity) and (2) to the evolution of the tumor (resistance or progression). These two factors are of central concern. In accordance with Shrestha et al. (2018), oncologists have to adjust the treatment to keep a balance between the quantity of life and the quality of life.

Patients’ representations of illness are related to the social and psychological effects of illness. Treatments, that is, chemotherapy, occupy a more peripheral place. Therefore, we can say that for patients, the quality of life has a wider meaning, which is not restricted to physical aspects.

Moreover, pain and suffering are associated with different meanings among oncologists and patients. In accordance with SR Theory, we may hypothesize that this is due to training and experience, which differ between the two groups. Because of their training, oncologists refer to pain, whereas patients, on account of their experience of disease, refer to both pain and suffering, by connecting physical, psychological, and social dimensions. These results are consistent with the concept of “total pain” described by Clark (1999).

Among patients, pain is associated with the cancer diagnosis, the conduct of medical examinations, and the side effects of treatments. Suffering is connected to pain, but it extends well beyond pain because it refers to the fear that treatments may not work and to the sense of vulnerability to death. It is also connected to the social and psychological consequences of illness, that is, physical deterioration and social isolation.

For oncologists, pain remains a matter confined to the medical field, being viewed as a symptom, as a sign of disease, of the development of the disease, and of tolerance to treatment. Therefore, when an oncologist listens to a patient expressing his or her pain, they do so through this prism, seeking indicators in the patient’s discourse that will lead to recommending further examinations or adapting the treatment.

Thus, patients speak of their pain, as well as expressing their suffering, their fear that their treatment may not work, and their fear of death. Oncologists, by contrast, seek to offer therapeutic solutions, thereby expressing their fear of entering the realm of psychology—a realm in which they have no expertise. These two ways of approaching pain and the tendency to ignore suffering can lead to misunderstandings. For patients, the fear remains of not being heard and of not being acknowledged as social and psychological beings with experiences that go beyond pain. The central feeling experienced by patients can be illustrated with the following quote:

When you see yourself lying in bed, and you can’t move, somewhere at the back of your mind, you say to yourself ‘I’m going to die’ (…). I tell myself I still want to carry on living a little longer, even though there are times when I just want to die (…) When you end up bald. Psychologically, you’re suffering (…) you tell yourself you’re a monster (…). It’s not easy. When you get a minor infection… It becomes a big deal in your mind because you know your body may not have what it takes to continue fighting the disease. (Patient 18)

The typical response of oncologists can be summarized with the following quote:

Am I going to die? How much longer have I got left? These are questions you can’t always answer. Am I going to lose my hair? Am I going to feel pain? How am I going to die? I don’t have a crystal ball. I have to see whether or not there’s a good response to the treatments. (Oncologist 5)

Thus, faced with the patients’ questions and suffering, the oncologists feel powerless. They take refuge behind the protocols, the effects of treatments, and the adjustments they can make. This study shows that patients struggle to express their suffering because it reveals their vulnerability and their fear of physical deterioration and death. For oncologists, pain can be objectified and assigned causes (e.g., disease and associated treatments) and therapeutic solutions (painkillers). In other words, patients talk about their pain but also seek to express their suffering, while oncologists focus on pain and tend to separate themselves from the expression of suffering since it falls outside their area of expertise, thus confronting them with their own powerlessness.

This study highlights the difficulty of dialogue between oncologists and patients, which lies at the heart of their relationship. As noted by Basset (2010, p. 483): “While they may be able to relieve pain, oncologists are often powerless in the face of suffering. It is precisely this distinction between pain and suffering, between a technical response and concern for the other, that needs to be examined. The way in which oncologists respond to patients complaining of pain is a major challenge for the therapeutic relationship.”

Conclusion

This mixed methods study is based on a small sample of participants. Thus, no general conclusions can be drawn. Despite this limitation, it offers new perspectives on how patients could be encouraged to express their suffering during consultations with their oncologist and how oncologists could approach the suffering expressed by their patients.

First, an interview guide could be useful for oncologists to evaluate a level of alert concerning patients’ suffering. This interview guide could be used as a training tool to support oncologists in managing patients’ suffering. Second, the guide could also be used as part of a multidisciplinary approach combining the perspectives of oncologists, psychologists, and other health professionals as a basis for discussion within teams and could lead to the decision to prescribe psychological care and, if necessary, individual therapy.

The first part of the interview guide could include items concerning how the illness is affecting social life: that is, a global question like “How are things going with your relatives?” and “your work?”. The second part could comprise specific questions extracted from a validated scale like the Hospital Anxiety and Depression Scale (Zigmond & Snaith, 1983), used for measuring emotional distress in medical patients. This scale contains items concerning (1) anxiety, that is, “Do you sometimes feel tense or wound up?” “Do you sometimes get a sort of frightened feeling as if something awful is about to happen?” and (2) depression, that is, “Do you enjoy watching TV programs?”.

This interview guide could be useful in overcoming barriers to the active management of pain and suffering in oncology. On the one hand, it could help oncologists to address these questions during consultations; on the other hand, it could facilitate the expression of pain and suffering by patients. Of course, it would be necessary to build and test the tool to measure its practicality for oncologists and its acceptance by patients.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Canceropôle Grand-Est [grant number: 9210AAO001S03440, 2016] and the Fondation de France [grant number: Engt 00069105, 2016].

ORCID iD

Edith Salès-Wuillemin

Note

Appendix

References

Abreu Lopes

Gaskell

(2015). Social representations and societal psychology. In Sammut

Andreouli

Gaskell

Valsiner

(Eds.), The cambridge Handbook of social representations(cambridge handbooks in psychology (pp. 29–42). Cambridge University Press. 10.1017/CBO9781107323650.005

Abric

(2005). Méthodes d'étude des représentations sociales. ERES. 10.3917/eres.abric.2003.01

Basset

(2010). Regard éthique du médecin soignant... sur la douleur et la souffrance. Dans. In Hirsch

(Ed.), Traité de bioéthique: III - handicaps, vulnérabilités, situations extrêmes (pp. 480–490). ERES. 10.3917/eres.hirsc.2010.03.0480

Bosire

E. N.

Mendenhall

Weaver

L. J.

(2020). Comorbid suffering: Breast cancer survivors in South Africa. Qualitative Health Research, 30(6), 917–926. https://doi.org/10.1177/1049732320911365

Bouckenaere

(2007). La douleur chronique et la relation médecin-malade. Cahiers de psychologie clinique, n 28(1), 167–183. https://doi.org/10.3917/cpc.028.0167

Brennan

Carr

D. B.

Cousins

(2007). Pain management: A fundamental human right. Anesthesia and Analgesia, 105(1), 205–221. http://www.ethimedix.com/wp-content/uploads/2016/02/Pain-Management-a-Fundamental-Human-Right.pdf. https://doi.org/10.1213/01.ane.0000268145.52345.55

Clapin

(Ed.), (2002). Philosophy of mental representation. OUP.

Clark

(1999). ‘Total pain', disciplinary power and the body in the work of Cicely Saunders, 1958–1967. Social Science and Medicine, 49(6), 727–736. https://doi.org/10.1016/s0277-9536(99)00098-2

Dany

Dormieux

Futo

Favre

(2006). La souffrance: Représentations et enjeux. Recherche en soins infirmiers, N° 84(1), 91–104. https://doi.org/10.3917/rsi.084.0091

10.

Den Beuken-van Everdingen

Marieke

H. J.

Hochstenbach

L. M. J.

Joosten

E. A. J.

Tjan-Heijnen

V. C. G.

Janssen

D. J. A.

Janssen

D. J. A.

(2016). Update on prevalence of pain in patients with cancer: Systematic review and meta-analysis. Journal of Pain and Symptom Management, 51(6), 1070–1090.e9. http://doi.org/10.1016/j.jpainsymman.2015.12.340

11.

Frank

(2001). Can We Research Suffering? Qualitative Health Research. 11. 353–62. https://doi.org/10.1177/104973201129119154

12.

Fromage

Hatti

(2015). Accompagnement phénoménologique de l'expérience douloureuse en soins palliatifs. L’Encéphale, 171(4), 262–267. https://doi.org/10.1016/j.encep.2014.10.019

13.

Granek

Nakash

Shapira

Ariad

BenDavid

M.A.

(2020), Oncologists, oncology nurses and oncology social workers experiences with suicide: impact on patient care, Journal of Psychosocial Oncology, 1–14, https://doi.org/10.1080/07347332.2020.1755763

14.

Hubbard

T. L.

(2007). What is mental representation? And how does it relate to consciousness? Journal of Consciousness Studies, 14, 37–61.

15.

IRaMuTeQ 07 alpha (2020). Interface de R pour les Analyses Multidimensionnelles de Textes et de Questionnaires. http://www.iramuteq.org/documentation/fichiers/documentation_19_02_2014.pdf

16.

Jeoffrion

Dupont

Tripodi

Roland-Lévy

(2016). Représentations sociales de la maladie: Comparaison entre savoirs « experts » et savoirs « profanes. L'Encephale, 42(3), 226–233. https://doi.org/10.1016/j.encep.2015.12.007

17.

Krawczyk

Woods

Clark

(2018). Total pain: origins, current practice, and future directions, Omsorg: The Norwegian Journal of Palliative Care, 2, http://eprints.gla.ac.uk/167847/

18.

Makhlouf

S. M.

Pini

Ahmed

Bennett

M. I.

(2020). Managing pain in people with cancer–A systematic review of the attitudes and knowledge of professionals, patients, caregivers and public. Journal of Cancer Education: The Official Journal of the American Association for Cancer Education, 35(2), 214–240. http://doi.org/10.1007/s13187-019-01548-9

19.

Morant

(2006). Social representations and professional knowledge: The representation of mental illness among mental health practitioners. The British Journal of Social Psychology, 45(Pt 4), 817–838. https://doi.org/10.1348/014466605X81036

20.

Moscovici

(1984). The phenomenon of social representations. In Farr

Moscovici

(Eds.), Social representations (pp. 3–69). Cambridge University.

21.

Moscovici

(2000). Social representations: Explorations in social psychology. Cambridge University Press.

22.

Nakash

Liphshitz

Keinan-Boker

Levav

(2013). The effect of cancer on suicide among elderly Holocaust survivors. Suicide and Life-Threatening Behavior, 43(3), 290–295. https://doi.org/10.1111/sltb.12015

23.

Nakash

Shemesh

Nagar

Levav

(2012). Cancer and common mental disorders in the community: Results of the Israel-World Mental Health Survey. The European Journal of Psychiatry, 26(3), 174–184. https://doi.org/10.4321/S0213-61632012000300004

24.

Péoc’h

(2014). Représentations sociales de la douleur. Les dossiers des sciences de l’éducation, 31, 81–99. https://journals.openedition.org/dse/593

25.

Philippin

(2009). Thérapie cognitive comportementale en soins palliatifs. InfoKara, 24(1), 11–26. https://doi.org/10.3917/inka.091.0011

26.

Pinto

Bhola

Chandra

P. S.

(2019). End-of-life care is more than wound care”: Health-care providers’ perceptions of psychological and interpersonal needs of patients with terminal cancer. Indian Journal of Palliative Care, 25(3), 428–435. https://doi-org.nwulib.nwu.ac.za/10.4103/IJPC.IJPC_26_19

27.

Rodriguez

C. S.

Van Cott

M. L.

(2005). Speech impairment in the postoperative head and neck cancer patient: Nurses’ and patients’ perceptions. Qualitative Health Research, 15(7), 897–911. https://doi.org/10.1177/1049732305278903

28.

Rozmovits

Ziebland

(2004). What do patients with prostate or breast cancer want from an Internet site? A qualitative study of information needs. Patient education and counseling, 53 1, 57–64. https://doi.org/10.1016/S0738-3991(03)00116-2

29.

Zigmond

A.S.

Snaith

R.P.

(1983). The hospital anxiety and depression -scale, Acta Psych. Scand, 67, 361–370