Abstract
Nurses, nurse’s aides, and physicians were presented with vignettes describing elderly patients and were asked to assess their level of pain from four external cues (facial expression, verbalizations, avoidance of movements and positions, and interpersonal contact) in three conditions: when the illness was not known, when it was known to be arthritis, and when it was known to be cancer. For all health caregivers, the most important cue for judging pain was patients’ facial expression. When the nature of the illness was not known, the impact of this cue was stronger than when the nature of the illness was known.
Nurses play a key role in assessing patients’ pain. Even with the institutionalized use of pain scales, assessing pain remains a challenging issue: (a) many patients experience difficulty in expressing their pain level, (b) health caregivers and patients may not completely understand each other, (c) certain patients resign themselves to suffer and others exaggerate their pain, and (d) patients’ pain reports are typically reinterpreted in light of the diagnosis (Igier et al., 2007).
In encounters with patients, nurses are sensitive to external, easily available cues such as facial grimacing, maintenance of an abnormal body position, restriction of movements, complaints about pain, and signs of possible depression. Igier et al. (2007) studied the cognitive processes by which, using these pain behaviors, nurses and nurse’s aides assessed the levels of pain. They found that nurses and nurse’s aides combined the information using an additive cognitive operation, that is, that each cue had an independent effect.
This study complemented Igier et al.’s (2007) study in two ways. First, by incorporating the type of illness as a cue, it examined the impact of knowing or not knowing the nature of the patient’s illness on the way nurses judge pain. In professional life, nurses encounter, often in close succession, patients suffering from different illnesses. The no-known-diagnosis-condition is also a realistic category for practicing health-care professionals, and frequently, patients’ underlying diagnoses are unknown. Knowledge of the illness can change the effect of the other factors on pain judgment. For example, facial grimacing might have more impact in case of advanced cancer than in case of early-stage osteoarthritis. Second, by incorporating physicians as participants, this study aimed to find out whether they too would use an additive combination rule to judge pain.
Method
Participants
The sample was composed of 79 nurses (74 females and 5 males) aged 20–61 years (M = 35 years), 75 nurse’s aides (71 females and 4 males) aged 18–58 years (M = 37 years), and 36 physicians (6 females and 30 males) aged 31–62 years (M = 46 years). More than 90 percent of each category of participants worked in hospitals or health centers.
Material and procedure
The material consisted of 72 cards containing a story of a few lines (Anderson, 2008). The set of stories was composed by orthogonal combination of the levels of five factors: (a) patient’s illness (early osteoarthritis, advanced cancer, or unknown), (b) patient’s facial expression (her face “appears grimacing” or “relaxed”), (c) patient’s movements and positions (“does not seem to avoid certain movements or certain positions,” “movements are rare and slow but the patient does not seem to avoid certain positions,” or “movements and positions seem somewhat constrained”), (d) patient’s verbalization (“complains of pain” or “does not spontaneously complain”), and (e) patient’s willingness to make personal contact (“it does not seem difficult to establish contact with her” or “she seems to avoid relating to others”). The design was, consequently, a five within-subject factor design: Illness × Facial expression × Movements × Verbalizations × Contacts, 3 × 2 × 3 × 2 × 2.
All patients were aged between 75 and 80 years and were identified as “Mrs X.” Under each story, there were the question “What amount of pain does Mrs X seem to be feeling currently?” and a 19-cm response scale with a left-hand anchor of “No pain” and a right-hand anchor of “Very intense pain.” The cards were arranged by chance and in a different order for each participant. The site was a vacant room in the hospital or health center. Each person was tested individually. Each participant worked at his or her own pace. This study was approved by the Ethics and Work Laboratory of the Institute for Advanced Studies, Paris, France.
Results
For each scenario, the distance was measured between the left-hand anchor and the answer, and an analysis of variance (ANOVA) was conducted on these measures of distance. In light of the multiplicity of comparisons, the level of significance was set at .001.
The between-participants factor—type of health professional—was not significant, that is, there were no significant differences between the mean ratings of nurses, nurse’s aides, and physicians. Each of the five within-subject factors, however, had a significant effect. When the patient’s movements and positions were restricted, the judged pain level was higher (M = 10.77) than in the other case (M = 8.12), F(2,368) = 253.93, η2p=.58. When the patient’s face was grimacing, the judged pain level was higher (M = 10.49) than when it was relaxed (M = 8.06), F(1,184) = 405.18, η2p=.69. When interpersonal contact was easy, the judged pain level was lower (M = 8.36) than in the other case (M = 10.19), F(1,184) = 226.92, η2p=.55. When the patient was complaining of pain, the judged pain level was higher (M = 9.98) than when the patient was not complaining (M = 8.57), F(1,184) = 160.55, η2p=.47. Finally, when the patient was suffering from cancer, the judged pain level was higher (M = 10.27) than when the illness was not indicated (M = 8.57) or when it was osteoarthritis (M = 8.99), F(2,368) = 40.88, η2p=.18. The Illness × Facial expression was significant (Figure 1). When the illness was not known, the face effect was stronger than when the illness was known, F(2,368) = 13.99, η2p=.07.
Effect of ease of contact, grimacing, and known or unknown diagnosis on pain judgment. The judged pain levels are on the vertical axis. The three levels of the movement and position factors are on the horizontal axis. Each curve corresponds to one level of the grimacing factor. Each panel corresponds to one modality of the diagnosis factor.
Discussion
Knowledge of the illness had an impact, but it was very limited. When the nature of the illness was not known, the impact of facial expression on pain judgment was slightly stronger than in the other cases. The impact of the other cues (e.g. verbalizations) was not affected by the knowledge of the illness. All participants, however, judged the cancer patient’s level of pain as higher than the osteoarthritis patient’s level of pain. The physicians judged patients’ pain using the same kind of additive-type rule as that used by the nurses and nurse’s aides. These findings are consistent with (a) suggestions by Craig et al. (2011) that facial expressions provide the most sensitive information about pain level; (b) findings by Marquié et al. (2008) that physicians rated pain by additively combining patient’s pain rating, behavioral manifestations of pain, and signs of the severity of the pathology; and (c) findings by Chibnall et al. (1997), who showed that medical students combined the reported pain intensity and the medical evidence in an additive fashion.
The use of a largely additive rule by physicians and nurses is striking because it was not evident, on a priori grounds, that they would do this. More sophisticated rules, involving many interactions of the kind that was observed for illness and facial grimacing, could have been found. For example, the joint presence of verbal complaints and lack of contact could have had more impact on pain judgment than the sum of the separate impact of these two cues.
The ability to be sensitive to the acute pain of others and to assess its degree is, no doubt, of sufficient importance to humans in society that it has become part of our common evolutionary heritage. It is not surprising, therefore, that nurses, nurse’s aides, and physicians use the same cognitive schema in judging patients’ pain. This shared schema greatly facilitates communication between them about pain.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.