C 2006) Journal of Behavioral Medicine, Vol. 29, No. 2, April 2006 ( DOI: 10.1007/s10865-005-9042-3
Factors Affecting Hospital Staff Judgments About Sickle Cell Disease Pain James Elander,1,6 Malgorzata Marczewska,2 Roger Amos,3 Aldine Thomas,4 and Sekayi Tangayi5 Accepted for publication: November 2, 2005 Published online: February 22, 2006
Judgments about people with pain are influenced by contextual factors that can lead to stigmatization of patients who present in certain ways. Misplaced staff perceptions of addiction may contribute to this, because certain pain behaviors superficially resemble symptoms of analgesic addiction. We used a vignette study to examine hospital staff judgments about patients with genuine symptoms of analgesic addiction and those with pain behaviors that merely resemble those symptoms. Nurses and doctors at hospitals in London, UK, judged the level of pain, the likelihood of addiction, and the analgesic needs of fictitious sickle cell disease patients. The patient descriptions included systematic variations to test the effects of genuine addiction, pain behaviors resembling addiction, and disputes with staff, which all significantly increased estimates of addiction likelihood and significantly decreased estimates of analgesic needs. Participants differentiated genuine addiction from pain behaviors resembling addiction when making judgments about addiction likelihood but not when making judgments about analgesic needs. The treatment by staff of certain pain behaviors as symptoms of analgesic addiction is therefore a likely contributory cause of inadequate or problematic hospital pain management. The findings also show what a complex task it is for hospital staff to make sensitive judgments that incorporate multiple aspects of patients and their pain. There are implications for staff training, patient education, and further research. KEY WORDS: vignette study; pain judgments; addiction; pseudoaddiction; analgesics; sickle cell disease.
INTRODUCTION
In vignette studies, participants make judgments about fictitious patients. The approach is inevitably artificial but allows controlled experiments in which specific factors are systematically manipulated. Previous vignette studies focused on chronic pain conditions, such as low back pain, and required participants to rate the levels of pain, emotional distress, and disability experienced by the ‘person in pain’. Studies with student participants found that ratings were higher when the cause of pain was outside the control of the patient, when there was medical evidence in support of the pain, and when the relationship with the person in pain was positive rather than adversarial (Tait and Chibnall, 1994). Those factors also affected ratings of the person in pain’s character and judgment (Chibnall and Tait, 1995). A conclusion was that ‘chronic pain
There is greater scope for contextual influences in the treatment of pain than in almost any other area of medicine, and pain is frequently under-treated, especially among ethnic minority patients. Research on contextual influences can contribute to better and more equitable pain management, as well as testing psychological theories of helping behavior. 1 Thames
Valley University, London, UK. Primary Care Trust, London, UK. 3 Homerton University Hospital, London, UK. 4 Barts and the Royal London Hospital, London, UK. 5 Newham University Hospital, London, UK. 6 To whom correspondence should be addressed at Psychology Department, Thames Valley University, St Mary’s Road, London W5 5RF, UK; e-mail:
[email protected]. 2 Newham
203 C 2006 Springer Science+Business Media, Inc. 0160-7715/06/0400-0203/0
204 patients can be stigmatized by contextual factors’, especially those who ‘present in an adversarial manner, complain of severe pain in the absence of objective medical findings, and who in some way contributed to their injury’ (Chibnall and Tait, 1995, p. 437). The main findings were replicated in samples of medical students (Chibnall et al., 1997), medical centre employees (Chibnall and Tait, 1999), and hospital physicians (Tait and Chibnall, 1997). Attribution theory provides a framework for understanding findings like those. In Weiner’s (1980) model, helping behavior is affected by feelings of sympathy or anger, which themselves are affected by attributions about the victim’s responsibility for events (see also Schmidt and Weiner, 1988). Responsibility for events has two aspects: onset controllability refers to the victim’s control over the causes of the problem, and maintenance controllability refers to current efforts to control the problem. The effects of attributions about both onset and maintenance controllability have been demonstrated in medical contexts. One study showed that medical students were more willing to prescribe tranquillizers or antidepressants to help hypothetical patients cope with less controllable rather than more controllable life events (Brewin, 1984). Another showed that doctors’ and nurses’ attitudes towards fictitious patients were affected by whether the patients had behaved in ways that could have reduced the risk of illness (Marteau and Riordan, 1992). Another showed that nurses rated patients who coped poorly with pain following minor abdominal surgery as more demanding, dependent, and unpopular (Salmon and Manyande, 1996). Most recently, students rated fictitious patients with shoulder pain more sympathetically if the patients had followed medical advice and coped with pain. Students with an unsupportive attributional style were more likely to select a comfortable treatment option for patients who had followed advice and coped with pain, and an uncomfortable one for patients who had not (Lundquist et al., 2002). To our knowledge, however, no controlled studies of biases in pain management have been conducted in the context of chronic, life threatening medical conditions that cause episodic or chronic pain, such as cancer, HIV/AIDS and sickle cell disease. In conditions like those, assessment of analgesic addiction is problematic (Kirsh et al., 2002), patients can be stigmatized as drug addicts (Hung et al., 2001; Maxwell et al., 1999), and there is considerable evidence of under-treatment of pain (Drayer et al., 1999), which can lead to ‘pseudoaddiction’ when
Elander, Marczewska, Amos, Thomas, and Tangayi patients resort to exaggerated or manipulative pain behaviors that are perceived by staff as drug seeking or addictive behaviors (Weissman and Haddox, 1989). Sickle cell disease is an inherited blood disorder in which recurrent episodes of severe ischemic pain occur when sickled red blood cells cause tissue infarction. The most severe painful episodes are treated in hospital with injections or infusions of morphine, diamorphine or meperidine (pethidine). Painful episodes are the most common reason for hospital attendance in sickle cell disease, yet pain management remains one of the most problematic and controversial aspects of treatment for sickle cell disease, and interpersonal as well as pharmacological factors affect the quality of pain management (Elander and Midence, 1996). In Europe and North America, sickle cell disease patients are almost all members of ethnic minorities, which is a risk factor for poor pain management across a range of conditions (Green et al., 2003; Ng et al., 1996; Todd et al., 2000). Symptoms of sickle cell disease usually begin in childhood, so most adult patients have substantial experience with pain coping and attend hospital only when in very severe pain. Patients sometimes report being unjustly suspected or accused of addiction (Harris et al., 1998; Maxwell et al., 1999; Shelley et al., 1994), and surveys of hospital staff show that many overestimate the prevalence of addiction among sickle cell disease patients (PackMabien et al., 2001; Shapiro et al., 1997; Waldrop and Mandry, 1995). A recent study of pain behaviors and analgesic use among people with sickle cell disease showed that many attempts to control pain superficially resembled symptoms of addiction, making patients vulnerable to misperceptions of addiction. Genuine symptoms of addiction took place in the absence of pain or involved analgesics used for reasons other than for pain control, such as to alter mood, whereas pain behaviors that merely resembled addiction involved the same behaviors but took place in the presence of pain or attempts to control pain (Elander et al., 2003). Examples corresponding to the DSMIV symptoms of substance dependence (American Psychiatric Association, 2000) are given in Table I. Genuine symptoms were correlated with factors previously associated with analgesic addiction, whereas pain behaviors resembling addiction were correlated with factors previously associated with pseudoaddiction (Lusher et al., in press). Problematic outcomes of hospital pain management, of which disputes
Judgments About Sickle Cell Disease Pain
205
Table I. Pain Behaviors Resembling Addiction and Genuine Symptoms of Analgesic Addiction (Adapted from Lusher et al., in press; see also Elander et al., 2003) DSM-IV symptom Tolerance
Withdrawal symptoms
Greater use than intended
Attempt to give up or cut down
Excessive time spent
Social impairment
Use despite known problems
Cravingb
Pain behavior resembling addictiona
Genuine symptom of analgesic addictiona
Larger or more frequent doses needed to control Larger or more frequent doses taken to obtain worsening pain. euphoria, relaxation, or effects other rather than pain relief. Pain returns or intensifies on discontinuation of Classic opiate withdrawal symptoms (e.g. analgesics. insomnia, nausea, sweating, cramps), or analgesics are used to avoid or treat withdrawal symptoms. Analgesics are taken more often or in larger Analgesics are taken more frequently or in doses because of associated euphoria, larger doses than intended because pain is relaxation, or other changes in mood. more severe or more prolonged than usual. Attempts to give up or cut down because of Attempts to cut down or give up analgesics concerns about becoming addicted or because of concerns about their effectiveness consequences of analgesic use unrelated to to control pain. pain. Time spent obtaining, using and recovering from Time spent obtaining, using and recovering from analgesics in the absence of pain or in analgesics is associated with pain and attempts attempts to obtain euphoria, relaxation or to control pain. changes in mood. Analgesics used to control pain lead to missed Activities with friends, family or at work are activities with friends, family or at work. missed because of analgesic use to obtain euphoria, relaxation or changes in mood. Analgesics continue to be used for pain relief Analgesics continue to be used despite despite recognized side effects. associated social or psychological problems, or continue to be used other than for pain control despite problems with analgesia. Desire to take analgesics consists mainly of Desire to take analgesics consists of seeking seeking relief from pain. euphoria, relaxation, or changes in mood.
a Elander
et al. (2003) and Lusher et al. (in press) refer to pain behaviors resembling addiction as ‘pain-related symptoms’ and genuine symptoms of addiction as ‘non-pain-related symptoms’. b Craving is not a DSM-IV symptom but was included in the previous research on pain behaviors and symptoms of addiction.
between patients and staff were the most common, were associated with pain behaviors resembling addiction rather than genuine symptoms of addiction, suggesting that the misperception by staff of analgesic addiction has a more important influence on problematic pain management than genuine analgesic addiction (Elander et al., 2004). Perceived analgesic addiction may be treated by staff as evidence that patients have failed to cope with their condition and are responsible to a greater degree for the maintenance of their pain, leading to lower levels of sympathy and less sympathetic treatment decisions. There is so far no direct evidence, however, about the perception by hospital staff of genuine analgesic addiction or pain behaviors that resemble addiction. In the present study we tested the effects of both types of behavior on hospital staff judgments about fictitious patients with sickle cell disease. Participants made judgments about the level of pain, the likelihood of addiction, and the analgesic needs of patients described in vignettes that incorporated several manipulations.
First, the vignettes included one of three types of statement about analgesic use, indicating a) no problematic analgesic use, b) pain behaviors resembling addiction, or c) genuine symptoms of analgesic addiction. The statements were based directly on descriptions from patient interviews in a previous study of analgesic use among people with sickle cell disease (Elander et al., 2003). Second, the vignettes included one of two types of statement indicating that patients had either a) co-operated with treatment or b) engaged in disputes with staff about pain management. Again, the statements were based directly on descriptions from patient interviews in a previous study of problematic hospital management of sickle cell disease pain (Elander et al., 2004). Third, the vignettes described patients with histories of either a) occasional or b) frequent hospital admissions for pain. Frequently hospitalized sickle cell disease patients are sometimes regarded as distinct from other sickle cell disease patients and may be treated differently by hospital staff. For example, frequently admitted sickle cell disease patients
206
Elander, Marczewska, Amos, Thomas, and Tangayi
reported adopting different strategies to negotiate their pain management (Maxwell et al., 1999), and hospital staff provided lower analgesic doses for children with histories of frequent hospitalizations for sickle cell disease pain (Armstrong et al., 1992). The aim of the study was to identify factors that could influence pain management by affecting staff judgments. We wished to test two hypotheses. First, we predicted that analgesic use, disputes with staff, and frequent hospitalizations would all give rise to less favorable judgments, that is, lower estimates of pain, higher estimates of addiction likelihood, and lower estimates of analgesic needs. Second, we predicted that, for analgesic use, genuine symptoms of addiction and pain behaviors resembling addiction would both give rise to less favorable judgments than no problematic analgesic use, but would not differ significantly from one another. We also tested the effects of participant factors, including age, gender, physicians versus nurses, specific professional training, and frequency of contact with sickle cell disease patients. METHOD
hospital] for treatment of painful sickling episodes. [No disputes statement/disputes statement]. When s/he was asked about the strong (opiate) painkillers s/he is prescribed for use at home, s/he said that [no problematic analgesic use statement/pain behaviors resembling addiction statement/genuine symptoms of addiction statement].
The statements themselves are given in the appendix. These were individualized to present each vignette as a plausible individual patient description and minimize the scope for hypothesis guessing. The statements were derived from interviews with sickle cell disease patients and selected to meet previously reported definitions of pain behaviors resembling addiction, genuine symptoms of addiction, and disputes with staff (Elander et al., 2003, 2004). Ratings for each vignette were made using sevenpoint Likert type scales. The vignettes and ratings scales were presented in the form of a questionnaire asking firstly for information about participants and then presenting each of the 12 vignettes with three ratings scales for each. Vignettes were compiled in a random order, and two versions of the questionnaire were produced, with one presenting the vignettes in the reverse order compared with the other.
Design Participants and Procedure Vignette descriptions of 12 fictitious patients were manipulated in a three-factor (3 × 2 × 2) within-subjects design. The independent variables were analgesic use (no problematic use vs. pain behaviors resembling addiction vs. genuine symptoms of addiction), disputes (absent vs. present), and hospitalizations (occasional vs. frequent). The dependent variables were ratings for each fictitious patient to indicate judgments about the degree of pain, the likelihood of addiction to analgesics, and the priority that should be given to analgesia. Materials The vignettes included statements representing the conditions of the independent variables. There were six male and six female ‘patients,’ with equal numbers of males and females for each level of each independent variable. The vignettes were in the form: . . . . is a young [man/woman] with sickle cell disease who [comes to hospital very rarely/is frequently in
Participants were recruited from the clinical staff at three east London hospitals. The inclusion criteria were working full time in parts of the hospitals where patients with sickle cell disease were routinely treated. Questionnaires were distributed to staff working in inpatient wards, emergency departments, maternity wards, and medical admissions units. Of 78 who were invited to participate, 59 returned completed questionnaires (30 completed the version with the vignettes in one order and 29 the version with the order reversed), giving a response rate of 76%. There were 46 (78%) females and 13 (22%) males. There were 11 (19%) doctors, 46 (78%) nurses, and two (3%) health care assistants. Twenty participants (35%) were aged 18–30 years, 25 (44%) aged 31–40, and 12 (21%) aged over 40 (two participants did not give their age). Fifteen participants (25%) had contact with sickle cell disease patients every day, 19 (32%) every week, 14 (24%) every month, and 11 (19%) less than every month. Thirty-three participants (56%) reported receiving training in pain management, 21 (36%) training
Judgments About Sickle Cell Disease Pain
207
Table II. F Ratios (Degrees of Freedom) from Within-Subjects Analysis of Variance Univariate tests Multivariate tests Analgesic use Disputes Hospitalization Analgesic use × disputes Analgesic use × hospitalization Dispute × hospitalization Analgesic use × disputes × hospitalization
(6, 53)∗∗∗
14.6 9.0 (3, 56)∗∗∗ 1.5 (3, 56) 9.3 (6, 53)∗∗∗ 1.7 (6, 53) 3.2 (3, 56)∗ 6.1 (6, 53)∗∗∗
Pain (1, 58)a
0.5 2.1 (1, 58) 0.3 (1, 58) 0.4 (1, 58)a 2.9 (1, 58) 10.0 (1, 58)∗∗ 9.5 (2, 116)∗∗∗
Addiction likelihood Analgesic needs 68.0 (1, 58)a∗∗∗ 21.9 (1, 58)∗∗∗ 1.5 (1, 58) 16.5 (2, 116)∗∗∗ 0.3 (1, 58) 0.1 (1, 58) 4.4 (2, 116)∗
8.0 (1, 58)a∗ 4.0 (1, 58)∗ 1.8 (1, 58) 0.7 (2, 116) 1.7 (1, 58) 3.1 (1, 58) 10.6 (2, 116)∗∗∗
a Degrees ∗p
of freedom adjusted from 2,116 to 1,58, using the ‘lower-bound’ epsilon (0.5). ≤ .05, ∗∗ p ≤ .01, ∗∗∗ p ≤ .001.
in hemoglobinopathies, and 6 (10%) training in substance dependence (some participants reported more than one type of training).
Data Analysis Repeated measures analysis of variance (ANOVA) was used to test within-subjects factors (the vignette manipulations), with between-subjects tests of differences between groups of participants, and tests of interactions between within-subject and between-subject factors. In each case we first tested multivariate effects (where the three types of ratings are combined), and examined univariate effects only in the presence of significant multivariate effects. For analgesic use, planned (simple) contrasts were used to test the differences between vignettes with no problematic analgesic use, pain behaviors resembling addiction, and genuine symptoms of addiction. For within-subjects effects with more than one degree of freedom (analgesic use and interactions with analgesic use), the most conservative, ‘lower-bound,’ estimate of epsilon was used to adjust degrees of freedom where Mauchly’s test of sphericity was significant. The first hypothesis was tested by the withinsubjects main effects, which we predicted would show that analgesic use, disputes, and frequent hospitalizations would all lower ratings of pain and analgesic needs, and increase ratings of addiction likelihood. The second was tested by the contrasts for analgesic use, which we predicted would show no significant differences between pain behaviors resembling addiction and genuine symptoms of addiction, with both differing significantly from no problematic analgesic use.
RESULTS Vignette Manipulations The results of the within-subjects analysis are given in Table II. There were significant multivariate effects of analgesic use, disputes, analgesic use × disputes, disputes × hospitalizations, and analgesic use × disputes × hospitalizations. The univariate tests showed no significant main effects on pain ratings, but, consistent with hypothesis one, both analgesic use and disputes significantly increased ratings of addiction likelihood and significantly decreased ratings of analgesic needs. Mean ratings for vignettes with different types of analgesic use are shown in Fig. 1. For ratings of addiction likelihood, all three types of vignette were significantly different from one another, including those with pain behaviors resembling addiction and genuine addiction (no problematic analgesic use vs. pain behaviors resembling addiction, F(1,58) = 69.4, p < .001; no problematic analgesic use vs. genuine addiction, F(1,58) = 82.9, p < .001; pain behaviors resembling addiction vs. genuine addiction, F(1,58) = 13.0, p = .001). For ratings of analgesic needs, there were significant differences between vignettes with no problematic analgesic use and both other types of vignette, but no significant difference between vignettes with pain behaviors resembling addiction and those with genuine addiction (no problematic analgesic use vs. pain behaviors resembling addiction, F(1,58) = 14.1, p < .001; no problematic analgesic use vs. genuine addiction, F(1,58) = 7.7, p = .008; pain behaviors resembling addiction vs. genuine addiction, F(1,58) = 0.1, p = .8). The significant two-way interactions are shown in Figs. 2 and 3. Figure 2 shows that hospitalization
208
Elander, Marczewska, Amos, Thomas, and Tangayi
Fig. 1. Mean addiction likelihood and analgesic needs ratings for vignettes with different analgesic use.
frequency moderated the effect of disputes on pain ratings. Disputes decreased pain ratings in the context of occasional hospitalizations, but had no effect when hospitalizations were frequent. Figure 3 shows that disputes moderated perceptions of addiction likelihood. Participants differentiated between genuine addiction and pain behaviors resembling addiction only in the presence of disputes; in the absence of disputes, ratings of addiction likelihood did not differ between vignettes with genuine addiction and those with pain behaviors resembling addiction. There were also significant 3-way interaction effects on all dependent variables, but these were much more difficult to interpret meaningfully. Mean values are given in Table III. The interaction for rat-
ings of pain appeared to show that genuine addiction and pain behaviors resembling addiction both decreased pain ratings when there were disputes, but only for vignettes with occasional hospitalizations, whereas the pattern was approximately reversed for vignettes with frequent hospitalizations. The interaction for ratings of addiction likelihood appeared to show that the trend for genuine addiction and pain behaviors resembling addiction to be differentiated only when there were disputes (as shown in Fig. 3) was more marked for vignettes with frequent hospitalizations. The interaction for ratings of analgesic needs appeared to show that pain behaviors resembling addiction decreased perceived analgesic needs when there were disputes, but only for vignettes with occasional hospitalizations, whereas the pattern was
Fig. 2.
Fig. 3.
Mean pain ratings for vignettes with occasional and frequent hospitalizations.
Mean addiction likelihood ratings for vignettes with different analgesic use.
Judgments About Sickle Cell Disease Pain
209
Table III. Mean (95% CIs) Ratings for Each Type of Vignette Rating
No problematic analgesic use No disputes Occasional hospitalizations Frequent hospitalizations Disputes Occasional hospitalizations Frequent hospitalizations Pain behaviors resembling addiction No disputes Occasional hospitalizations Frequent hospitalizations Disputes Occasional hospitalizations Frequent hospitalizations Genuine addiction No disputes Occasional hospitalizations Frequent hospitalizations Disputes Occasional hospitalizations Frequent hospitalizations
Pain
Addiction likelihood
Analgesic needs
4.6 (4.2–5.1) 5.2 (4.8–5.6)
2.6 (2.2–3.0) 2.8 (2.4–3.2)
4.9 (4.4–5.4) 5.0 (4.5–5.4)
4.9 (4.5–5.3) 4.8 (4.3–5.2)
3.9 (3.4–4.4) 3.8 (3.3–4.3)
5.0 (4.6–5.4) 4.5 (4.1–5.0)
5.1 (4.8–5.5) 4.6 (4.2–5.1)
5.0 (4.6–5.5) 5.3 (4.9–5.8)
4.8 (4.4–5.3) 4.3 (3.8–4.7)
4.3 (3.9–4.8) 5.2 (4.9–5.6)
4.9 (4.4–5.4) 5.0 (4.5–5.5)
3.8 (3.3–4.4) 4.6 (4.2–5.1)
5.1 (4.7–5.5) 4.7 (4.2–5.1)
5.4 (5.0–5.8) 5.1 (4.6–5.6)
4.6 (4.1–5.0) 4.3 (3.8–4.7)
4.6 (4.1–5.1) 4.6 (4.2–5.0)
5.6 (5.2–6.0) 6.1 (5.7–6.4)
4.5 (4.0–5.0) 4.4 (3.9–4.8)
approximately reversed for vignettes with frequent hospitalizations. Between-Subject Factors The between-subjects analyses revealed no significant main or interaction effects of hospital, doctors versus nurses/health care assistants, gender, or having received training in pain management, hemoglobinopathies, or substance dependence. For frequency of contact with sickle cell disease patients, there was a significant multivariate main effect but no significant interactions with any of the within-subjects factors. The univariate tests showed a significant main effect only for ratings of analgesic needs (F(1,57) = 6.3, p = .02). Participants with more frequent contact with sickle cell disease patients rated analgesic needs higher than did those with less frequent contact. The mean ratings were 4.7 (95% CI 4.4 – 5.1) for participants who saw sickle cell disease patients at least once a month, compared with 3.7 (95% CI 2.9 – 4.5) for those who saw sickle cell disease patients less than once a month. For age, there was no significant multivariate main effect but a significant age × analgesic use interaction. The univariate tests showed significant interactions for ratings of pain (F(4,108) = 3.4, p =
.01) and analgesic needs (F(4,108) = 4.0, p = .005). Figures 4 and 5 show that older participants were more sensitive to analgesic use; they gave higher pain and analgesic needs ratings for vignettes with no problematic analgesic use, and lower ratings for vignettes with genuine addiction and those with pain behaviors resembling addiction. DISCUSSION The results show that staff judgments about patients with sickle cell disease can be influenced by
Fig. 4. Mean pain ratings for vignettes with different analgesic use, made by participants of different ages.
210
Fig. 5. Mean analgesic needs ratings for vignettes with different analgesic use, made by participants of different ages.
contextual factors in ways that help to explain some of the difficulties experienced in hospital management of sickle cell disease pain. The first hypothesis was largely supported by the tests of main effects, in that analgesic use and disputes increased estimates of the likelihood of addiction and reduced estimates of analgesic needs, despite having no main effects on ratings of pain. The second hypothesis was also partly supported by the tests of contrasts. These showed that vignettes with genuine addiction and those with pain behaviors resembling addiction increased estimates of addiction likelihood and reduced estimates of analgesic needs, by comparison with vignettes with no problematic analgesic use. In judgments about analgesic needs, there was no difference between vignettes with genuine addiction and those with pain behaviors resembling addiction. The findings showed that the distinction between genuine symptoms of addiction and pain behaviors resembling those symptoms was not simply too fine or too subtle, for they were differentiated in judgments about the likelihood of addiction. However, they were not differentiated in judgments about analgesic needs, which is the judgment with most immediate impact on the treatment of pain. When judging analgesic needs, participants responded to patients with pain behaviors resembling addiction in the same way as to patients with genuine addiction, judging that their analgesic needs were lower than for patients with no signs of problematic analgesic use. Translated into clinical practice, patients who coped with pain in ways that resembled symptoms of addiction would receive less powerful analgesics, or lower or less frequent doses, exacerbating under-treatment of pain and prompting further attempts to cope that
Elander, Marczewska, Amos, Thomas, and Tangayi would be liable to misinterpretation by staff as symptoms of addiction. The interaction effects showed that judgments were affected by patient characteristics in combination. For example, when participants judged the likelihood that patients were addicted, they differentiated genuine symptoms of addiction from pain behaviors resembling addiction only in the presence of disputes, and the impact of disputes was greater when patients were frequently hospitalized. Making sensitive judgments about the treatment of pain that take account of multiple patient attributes is a complex task, and this was borne out to some extent by the between-subjects analyses. Those showed no effects of being a doctor rather than a nurse or healthcare assistant, or having specific training, but participants with more frequent contact with sickle cell disease patients made higher estimates of analgesic needs, suggesting that sensitivity to patients’ needs requires practical experience rather than specific formal knowledge or skills. Older participants, meanwhile, were no more likely than younger participants to differentiate genuine symptoms of addiction from pain behaviors resembling addiction, but appeared to be more sensitive to either type of analgesic use when estimating pain and analgesic needs. Compared with younger participants, they gave higher ratings for vignettes with no problematic analgesic use and lower ratings for those with genuine symptoms of addiction and pain behaviors resembling addiction. The judgments of the older participants may reflect more conservative attitudes to pain management in the context of the possibility of addiction, whereas those of younger participants may reflect a wider trend towards more liberal attitudes to analgesic prescribing. In a recent survey of pain physicians, for example, ratings of the potential for detrimental effects of strong opiates were lower than in previous surveys (Harden et al., 2005). The findings have implications for interventions to improve the assessment and treatment of pain. They support staff training with less emphasis on formal knowledge and more emphasis on practical experience with making assessments of patients who use analgesics in different ways and have disputes with staff. Interventions could also encourage staff to recognize that the distinction between genuine symptoms of addiction and pain behaviors that merely resemble addiction should be taken into account in judgments about analgesic needs as well as the likelihood of addiction. There may also be implications for patient education, to encourage patients to
Judgments About Sickle Cell Disease Pain communicate their pain in ways that are less likely to be misconstrued by staff as signs of addiction. Theoretical models exist for how observers decode pain behaviors to make judgments about patients’ pain (Prkachin and Craig, 1995; Prkachin et al., 1994), but not for how they make informal assessments of addiction, and the present findings are a small step towards such a model. They may also help to understand the behavioral mechanisms that give rise to pseudoaddiction, which occurs when patients in pain make attempts to obtain analgesia that are perceived by staff as drug seeking or addictive behaviors. Analysis of individual cases suggests that under-treatment of pain is the ultimate cause (Weissman and Haddox, 1989), but little is known about the specific ways that the development of pseudoaddiction is mediated by staff perceptions of, and attributions about, patient behaviors. Pain behaviors resembling symptoms of addiction may provide the first operational definition of behaviors that can be caused by attempts to cope with pain and are likely to be treated by observers as evidence of analgesic addiction. Vignette studies are not without limitations because the tasks presented to participants are necessarily artificial, but the method has been used successfully in several medical contexts, including the interpretation of physical symptoms (Swartzman and McDermid, 1993) and staff attitudes to patients (Marteau and Riordan, 1992), as well as the work on judgments about pain reviewed in the introduction. In the present study we attempted to minimize the artificiality of the task by basing the vignette descriptions on material from previous patient interview studies. An advantage of within-subjects designs, which have been used in a number of previous vignette studies of pain (e.g., Hadjistavropoulos et al., 1990; Tait and Chibnall, 1994), is that there is evidence of reasonable consistency within but not between individuals in medical staff judgments about pain (Chibnall et al., 2000). A potential disadvantage is the possibility of hypothesis guessing. We attempted to minimize this by individualizing the vignettes so they were presented as plausible individual cases, with a standard structure and form of presentation but without significant wording that was identical in more than one vignette. The use of two versions of the questionnaire, each presenting the vignettes in a different order, was designed to reduce the scope for order effects. There were no differences in ratings between the two orders of vignette presentation (multivariate F(3,55) = 0.47, p = 0.71), although
211 reversing the order of vignette presentation for half the participants does not fully control for order effects, which would require a Latin Square design in which each vignette occupied each place in the order. The findings could provide the basis for further research using different methods. One possibility is more naturalistic studies of staff responses to real patients whose pain behaviors and analgesic use have been assessed using the criteria that were employed in the design of the vignettes for the present study. Another is to develop and evaluate interventions to improve staff assessments of pain behaviors and symptoms of analgesic addiction. In both cases these could include studies of painful conditions other than sickle cell disease, for the present findings should be expected to generalize to other painful chronic conditions. ACKNOWLEDGMENTS Many thanks to the staff at Barts and the Royal London Hospital, Homerton University Hospital, and Newham University Hospital who took part in the study or helped to facilitate the data collection, especially Bernice Burton, John Coakley, Bunia Gorelick, Mahendra Kuruppurachch, Peter Loader, Joanne Morris, Rachel Simon, and Paul Telfer. Thanks also to Liz Robinson for the suggestion that led to the study, to David Hardman, Robin Iwanek, and Nick Troop for advice about the statistical analysis, and to the referees for their helpful comments on a previous draft. REFERENCES American Psychiatric Association (2000). DSM-IV-TR: Diagnostic and Statistical Manual of Mental Disorder, 4th ed., Washington DC: American Psychiatric Association. Armstrong, F. D., Pegelow, C. A., Gonzalez, J. C., and Martinez, A. (1992). Impact of children’s sickle cell history on nurse and physician ratings of pain and medication decisions. J. Pediatr. Psychol. 17: 651–664. Brewin, C. R. (1984). Perceived controllability of life-events and willingness to prescribe psychotropic drugs. Br. J. Soc. Psychol. 23: 285–287. Chibnall, J. T., and Tait, R. C. (1995). Observer perceptions of low back-pain: Effects of pain report and other contextual factors. J. Appl. Soc. Psychol. 25: 418–439. Chibnall, J. T., and Tait, R. C. (1999). Social and medical influences on attributions and evaluations of chronic pain. Psychol. Health 14: 719–729. Chibnall, J. T., Tait, R. C., and Ross, L. R. (1997). The effects of medical evidence and pain intensity on medical student judgments of chronic pain patients. J. Behav. Med. 20: 257–271. Chibnall, J. T., Dabney, A., and Tait, R. C. (2000). Internist judgments of chronic low back pain. Pain Med. 1: 231–237.
212 Drayer, R. A., Henderson, J., and Reidenberg, M. (1999). Barriers to better pain control in hospitalized patients. J. Pain Symptom Manage 17: 434–440. Elander, J., and Midence, K. (1996). A review of evidence about factors affecting quality of pain management in sickle cell disease. Clinical Journal of Pain 12: 180–193. Elander, J., Lusher, J., Bevan, D., and Telfer, P. (2003). Pain management and symptoms of substance dependence among patients with sickle cell disease. Soc. Sci. Med. 57: 1683– 1696. Elander, J., Lusher, J., Bevan, D., Telfer, P., and Burton, B. (2004). Understanding the causes of problematic pain management in sickle cell disease: Evidence that pseudoaddiction plays a more important role than genuine analgesic dependence. J. Pain Symptom. Manage. 27: 156–169. Green, C. R., Anderson, K. O., Baker, T. A., Campbell, L. C., Decker, S., Fillingim, R. B., Kaloukalani, D. A., Lasch, K. E., Myers, C., Tait, R. C., Todd, K. H., and Vallerand, A. H. (2003). The unequal burden of pain: Confronting racial and ethnic disparities in pain. Pain Med. 4: 277–294. Hadjistavropoulos, H. D., Ross, M. A., and von Baeyer, C. L. (1990). Are physicians’ ratings of pain affected by patients’ physical attractiveness? Soc. Sci. Med. 31: 69–72. Harden, R. N., Weinland, S. R., Remble, T. A., Houle, T. T., Colio, S., Steedman, S., and Kee, W. G. (2005). Medication quantification scale version III: Update in medication classes and revised detriment weights by survey of American Pain Society physicians. J. Pain 6: 364–371. Harris, A., Parker, N., and Barker, C. (1998). Adults with sickle cell disease: Psychological impact and experience of hospital services. Psychol. Health Med. 3: 171–179. Hung, C. I., Liu, C. Y., Chen, C. Y., Yang, C. H., and Yeh, E. K. (2001). Meperidine addiction or treatment frustration? Gen. Hosp. Psychiatry 23: 31–35. Kirsh, K. L., Whitcomb, L. A., Donaghy, K., and Passik, S. D. (2002). Abuse and addiction issues in medically ill patients with pain: Attempts at clarification of terms and empirical study. Clinical Journal of Pain 18: S52–S60. Lundquist, L. M., Higgins, N. C., and Prkachin, K. M. (2002). Accurate pain detection is not enough: Contextual and attributional style as biasing factors in patient evaluations and treatment choice. J. Appl. Biobehav. Res. 7: 114–132. Lusher, J., Elander, J., Bevan, D., Telfer, P., and Burton, B. (in press). Analgesic addiction and pseudoaddiction in painful chronic illness. Clinical Journal of Pain. Marteau, T. M., and Riordan, D. C. (1992). Staff attitudes towards patients: The influence of causal attributions for illness. Br. J. Clin. Psychol. 31: 107–110.
Elander, Marczewska, Amos, Thomas, and Tangayi Maxwell, K., Streetly, A., and Bevan, D. (1999). Experiences of hospital care and treatment for pain from sickle cell disease: Qualitative study. Br. Med. J. 318: 1585– 1590. Ng, B., Dimsdale, J. E., Rollink, J. D., and Shapiro, H. (1996). The effect of ethnicity on prescriptions for patient-controlled analgesia for postoperative pain. Pain 66: 9–12. Pack-Mabien, A., Labbe, E., Herbert, D., and Haynes, J. (2001). Nurses’ attitudes and practices in sickle cell pain management. Appl. Nurs. Res. 14: 187–192. Prkachin, K. M., Berzins, S., and Mercer, S. R. (1994). Encoding and decoding of pain expressions: A judgment study. Pain 58: 253–259. Prkachin, K. M., and Craig, K. D. (1995). Expressing pain: the communication and interpretation of facial pain signals. J. Nonverbal Behav. 19: 191–205. Salmon, P., and Manyande, A. (1996). Good patients cope with their pain: Post operative analgesia and nurses’ perceptions of their patients’ pain. Pain 68: 63–68. Schmidt, G., and Weiner, B. (1988). An attribution-affect-action theory of behavior—replications of judgments of help-giving. Pers. Soc. Psychol. Bull. 14: 610–621. Shapiro, B. S., Benjamin, L. J., Payne, R., and Heidrich, G. (1997). Sickle cell-related pain: perceptions of medical practitioners. J. Pain Symptom Manage. 14: 168–174. Shelley, B., Kramer, K. D., and Nash, K. B. (1994). Sickle cell mutual assistance groups and the health care delivery system. In Nash, K. B. (Ed.), Psychosocial Aspects of Sickle Cell Disease: Past Present and Future Directions of Research, New York: Haworth Press, pp. 243–259. Swartzman, L. C., and McDermid, A. J. (1993). The impact of contextual cues on the interpretation of and response to physical symptoms: A vignette approach. J. Behav. Med. 16: 183–198. Tait, R. C., and Chibnall, J. T. (1994). Observer perceptions of chronic low back pain. J. App. Soc. Psychol. 24: 415–431. Tait, R. C., and Chibnall, J. T. (1997). Physician judgments of chronic pain patients. Soc. Sci. Med. 45: 1199–1205. Todd, K. H., Deaton, C., D’Adamo, A. P., and Goe, L. (2000). Ethnicity and analgesic practice. Ann. Emerg. Med. 35: 11– 16. Waldrop, R. D., and Mandry, C. (1995). Health professional’s perceptions of opioid dependence among patients with pain. Am. J. Emerg. Med. 13: 529–531. Weiner, B. (1980). A cognitive (attribution) – emotion – action model of motivated behavior: An analysis of judgments of help giving. J. Per. Soc. Psychol. 39: 186–200. Weissman, D. E., and Haddox, J. D. (1989). Opioid pseudoaddiction: An iatrogenic syndrome. Pain 36: 363–366.
Judgments About Sickle Cell Disease Pain
213
APPENDIX: STATEMENTS USED TO CONSTRUCT THE VIGNETTES No Disputes 1. During those frequent hospital admissions his pain is managed without problems or arguments and he gets on well with the hospital staff. 2. On his last admission to hospital he got on well with the staff and was a very co-operative patient, even when in severe pain. 3. In hospital, she co-operates with all the staff in the management of her pain and her treatment is unproblematic despite being frequently in severe pain. 4. When he does go to hospital because of pain there are no problems or arguments with the staff and his treatment goes very smoothly despite being in great pain. 5. When she was last in hospital she got on well with the staff, and her pain was managed without arguments or problems. 6. She is regarded as a co-operative patient whose pain management is unproblematic in spite of the frequency and severity of her pain.
Disputes 1. On the last occasion there was an argument about her pain medication, and she got angry and upset because she thought she was being made to wait too long for pain killers, and screamed at the staff until she got her injections. 2. The last time he came to hospital he arrived screaming for painkillers and refusing to answer questions about his pain, saying “just give me my pain relief now!” He later made a complaint, saying that nurses had accused him of pretending to be in pain. 3. On the last occasion there was a major dispute on the ward that led to her discharge from hospital being brought forward. She had brought some painkilling medication from home into hospital and given some to other patients on the ward, and given some pain killing tablets that were dispensed for her in hospital to friends from the patient support group who had visited her in hospital. 4. Recently he had a dispute with the staff in hospital about what painkiller he should have. The doctor wanted to change his medication from pethidine injections to morphine tablets, but he didn’t think the tablets were working and wanted to go back to injections. There was an argument between him and the doctor about how his medication should be changed, and he refused to change medication as the doctor wanted him to and told the doctor to get lost. 5. There have been arguments in the hospital about her pain because the staff prefer her to have morphine for her pain but she believes she is allergic to morphine and insists on pethidine injections because that is how her pain was treated in the past. 6. The last time he was in hospital there was a dispute about the dose of analgesic he was given. He claimed that his dose had been reduced when he was transferred from casualty to the ward, and that the dose of diamorphine he was given on the ward was not enough to control his pain. He had looked at his chart and claimed that the nurse was giving him only half the dose he should have received, and accused her of cutting down the dose without the doctor’s permission.
No Problematic Analgesic Use 1. He had no problems with them. He never needed larger or more frequent doses, never used more than he intended, and never felt he should cut down or had allowed using painkillers to affect his family or social life. 2. There were no problems. She never used more and more tablets or took more than she had meant to, and had never felt that using the pain killers affected her social life or family, or that she should use less pain killers. 3. They did not cause any problems. She never had to take more than usual or more than she intended to take, never felt that she should use them less, and never felt that using painkillers affected her family or social life. 4. He did not have any problems with them, and never needed to take larger doses to control pain. He had never taken more of them than he should have, had never felt that he ought to use less or give them up, and the tablets did not affect his family or social life.
Pain Behaviors Resembling Addiction 1. He needed more and more to control the pain. Sometimes he felt as if his body was missing something when he didn’t take them when coming home from hospital after a painful episode. Sometimes he takes more than he should, thinking that the more he takes, the faster the pain will go, and although he has tried to stop using pain killers he finds that the pain means he soon has to go to the doctor for more. When he is in pain and has taken painkillers he often phones to say he won’t be able to come to work because he is feeling drowsy. 2. She needs to use larger amounts of painkillers to delay or avoid going to hospital when she is in pain. She experiences withdrawal symptoms when the pain subsides and she stops taking them, and she sometimes takes more than she intended to, when the pain does not respond to the tablets. She has tried to reduce how many tablets she takes, but needs them to control the pain, and also has to miss work sometimes because when she is in pain and takes painkillers, the tablets make her drowsy. 3. He now needs more of them to control the pain than previously, and sometimes he takes more than he intended when the tablets don’t seem to control the pain. He would like to cut down on the painkillers he takes, but has not been able to because of the pain he experiences. He also has to miss work sometimes when he has taken painkillers because they make him less alert even though he takes them only when in pain. 4. Sometimes she takes more pain killers than she should, when she is still in pain after taking pain killers, and has tried to give up using pain killers but when the pain returns she finds that nothing else works. She often misses work when she is in pain because the painkillers make her sleepy.
214
Elander, Marczewska, Amos, Thomas, and Tangayi
Genuine Symptoms of Addiction 1. She experiences withdrawal symptoms unless she increases the amount she takes, and is irritable and restless when she stops taking the tablets. When she is feeling low she sometimes takes more than she should. She wants to reduce the painkillers she uses because sometimes she avoids being with people when she has taken them, even when she is not in pain. 2. He has trouble relaxing when he is not in pain if he does not take them. He needs to take more of the tablets than in the past to avoid the agitated feeling he gets when he stops taking them, and sometimes he takes more than he should when he is feeling stressed, and when that happened once he stopped going to work for a short period. 3. He needed to take more and more to control the stomach cramps he experiences when withdrawing from pain killers, and was restless and agitated when he stopped taking them, even when not in pain. When feeling stressed he sometimes took more than he intended to, and stayed in and stopped going to work during a period when he was feeling depressed and taking a lot of pain killers. He has tried to cut down the painkillers because he thought he was taking too much and they were making him slow and sluggish. 4. She needed more and more to control the withdrawal symptoms that happened when she didn’t take them, and could not sleep properly without them, even when not in pain. She has to stop herself from getting carried away and taking too many pain killers when she is trying to make sure that the pain won’t come back, and has tried to wean herself off the tablets because she felt she was becoming immune to them. For a while she was taking them because she was feeling low, and stopped going out.