European Journal of Pain 11 (2007) 352–359 www.EuropeanJournalPain.com Oncologists’ perceptions of cancer pain management in Spain: The real and the ideal Joan Carulla Torrent a, Carlos Jara Sánchez b, Jaime Sanz Ortiz c, Norberto Batista López d, Carlos Camps Herrero e, Javier Cassinello Espinosa f, José Lizón Giner g, Joaquı´n Montalar Salcedo h, Kathryn Fitch Warner i,*, Milena Gobbo Montoya i, Eduardo Dı́az-Rubio Garcı́a j, on behalf of the ALGOS Group a i Medical Oncology Service, Hospital Universitari Vall d’Hebrón, Barcelona, Spain b Medical Oncology Service, Fundación Hospital Alcorcón, Madrid, Spain c Medical Oncology Service, Hospital Universitario Marqués de Valdecilla, Santander, Spain d Medical Oncology Service, Hospital Universitario de Canarias, Spain e Medical Oncology Service, Hospital General Universitario de Valencia, Spain f Medical Oncology Service, Hospital General Universitario de Guadalajara, Spain g Medical Oncology Service, Hospital Universitario de San Joan, Alicante, Spain h Medical Oncology Service, Hospital Universitario La Fe, Valencia, Spain Técnicas Avanzadas de Investigación en Servicios de Salud (TAISS), Calle Cambrils, 41-2, 28034 Madrid, Spain j Medical Oncology Service, Hospital Clı́nico Universitario San Carlos, Madrid, Spain Received 11 November 2005; received in revised form 6 April 2006; accepted 14 May 2006 Available online 11 July 2006 Abstract Aim: Studies in some countries suggest that cancer pain is often not adequately controlled, but little is known about the situation in Spain. The objective of this study was to identify medical oncologists’ perceptions about pain management in their patients. Methods: Two-round Delphi survey of 24 medical oncologists from 22 large, geographically diverse hospitals in Spain. Physicians rated each of 150 statements on a Likert scale (1 = strongly disagree; 5 = strongly agree). The mean, standard deviation and frequency of replies in three agreement categories were calculated for each item. Statements allowing comparison of oncologists’ perceptions of how pain is managed in routine clinical practice with how it should be managed were grouped together and analyzed. Results: The most notable discrepancies between the real and the ideal occurred in the failure to provide written information or to confirm that patients understand what they are told, the lack of comprehensive and systematic evaluation of pain, and the lack of use of non-pharmacological treatments (NPTs) for cancer pain. Conclusions: Medical oncologists need to improve their communication skills, providing patients with both written and verbal information about their disease and the plan for pain management. Pain should be evaluated at each patient visit using validated scales, and greater attention should be paid to the possible use of NPTs. Ó 2006 European Federation of Chapters of the International Association for the Study of Pain. Published by Elsevier Ltd. All rights reserved. Keywords: Pain control; Pain assessment; Physician preferences; Delphi * Corresponding author. Tel.: +34 91 731 0383; fax: +34 91 730 2893. E-mail address: kfitch@taiss.com (K. Fitch Warner). 1090-3801/$32 Ó 2006 European Federation of Chapters of the International Association for the Study of Pain. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ejpain.2006.05.006 J. Carulla Torrent et al. / European Journal of Pain 11 (2007) 352–359 1. Introduction Pain is one of the most frequently reported and feared symptoms in cancer patients. Estimates of the prevalence of pain from large, multicenter studies in different settings have varied from 57% of inpatients and outpatients (Larue et al., 1995), to 67% of outpatients with metastatic cancer (Cleeland et al., 1994), to over 70% of hospitalized patients with advanced cancer (Yun et al., 2003). Despite the widespread availability of safe and effective analgesics (Carr et al., 2004) and evidence-based guidelines for their use (Carr, 2001), large variability in how physicians manage cancer pain has been documented (Cleeland et al., 1994; Green et al., 2003; Von Roenn et al., 1993), and studies in many countries suggest that pain control is often inadequate (Di Maio et al., 2004; Larue et al., 1995; Vainio, 1995; Von Roenn et al., 1993; Zenz et al., 1995). Although cancer pain can be alleviated in approximately 90% of cases (Schug et al., 1990), it is estimated that about 20% of patients with cancer have uncontrolled pain (Ahmedzai, 1997), rising to almost 70% of those in their last year of life (Addington-Hall and McCarthy, 1995). Pain may be due to the location of the neoplasia, the effects of the treatment received, or other etiologies. Although the goal of therapy is to treat the source of pain, if this is not possible, the patient’s perception of pain can be altered in most cases using pharmacologic and non-pharmacologic approaches (Levy, 1996). Although many studies have been carried out in different countries on the prevalence of cancer pain and physician attitudes regarding its treatment, little literature on this subject exists with respect to Spain. Such studies as do exist have documented the prevalence of acute and chronic pain in the general population (Bassols et al., 1999; Catala et al., 2002) and in hospitalized patients (Canellas et al., 1993; Valero et al., 1995), but not in oncology patients. Likewise, few studies have been published regarding physician attitudes about pain treatment in Spanish patients (Farre et al., 1992; López de Maturana et al., 1993), and none of them have specifically explored medical oncologists’ attitudes and perceptions in this area. To fill this gap in knowledge, the ALGOS project was launched by a group of 58 medical oncologists from different centers in Spain in 1998. After holding a series of workshops to discuss the current situation of cancer pain management and to identify possible obstacles to adequate treatment, the ALGOS group designed a 3phase study to produce relevant and scientifically valid knowledge on the problem of pain and its management by medical oncologists in Spain. The objectives of this project are to identify the perceptions, attitudes and experiences of medical oncology specialists with regard to pain in their patients (phase 1); to study the association between pain, its management and patient characteristics 353 (phase 2); and to determine patient perceptions, attitudes and experiences with regard to pain (phase 3). The present study presents the partial results of the first phase of this project, comparing medical oncologists’ perceptions of how pain is managed in routine clinical practice (‘‘the real’’) and how it should be managed (‘‘the ideal’’). 2. Methods A 2-round Delphi survey of an expert panel of 24 medical oncologists in the ALGOS group was carried out between March and July 2004. The physicians included in the survey were from 22 mostly large teaching hospitals, representing 11 of Spain’s 17 Autonomous Communities, and were all chiefs of their respective Oncology Services or Units. Their mean age was 53 years (range 41–61). The survey instrument was a 150-item questionnaire developed by study investigators based on a literature review of the prevalence, evaluation and treatment of acute and chronic pain, clinical practice guidelines on pain management (Benedetti et al., 2000; Jacox et al., 1994), other documents of the ALGOS group, and previous questionnaires used to evaluate perceptions and attitudes of patients, their family members and health professionals in regard to post-surgical pain. The items were divided into six thematic areas related with cancer pain: information, evaluation, management, and the attitudes of patients, medical oncologists and non-oncologist physicians. Each item consisted of a statement – for example, ‘‘Medical oncologists provide their patients with sufficient information regarding the possible side effects of pain medications’’ – to be rated on a 5-point Likert scale, where 1 = strongly disagree and 5 = strongly agree. New questionnaires were prepared for the second round showing the mean and standard deviation (SD) together with the individual’s own rating for each statement, and panelists were asked to rate each item again. The mean and SD for each item in the second-round questionnaire were calculated, together with the proportion of experts who rated the item 1 or 2 (strongly or moderately disagree), 3 (neither agree nor disagree), and 4 or 5 (moderately or strongly agree). Of the 150 items, it was possible to group 69 of them into pairs (or occasionally groups of 3), resulting in 32 areas in which the physicians’ perceptions about routine clinical practice could be compared with what they considered desirable practice. All of these comparisons fell into the first three thematic areas: information, evaluation and management of cancer pain. 3. Results All 24 panelists answered the first round questionnaire, and 22 answered the second. There were no substantive changes in ratings between the two rounds. 354 J. Carulla Torrent et al. / European Journal of Pain 11 (2007) 352–359 Table 1 shows the panelists’ mean ratings for the seven comparisons that could be made in the area of pain information, with statements pertaining to the same general topic grouped together. All panelists agreed that the medical oncologist, rather than the nurse, is responsible for informing the patient about pain-related issues, and that this information is and should be provided directly to the patient, rather than through family members. It was also strongly agreed Table 1 Mean ratings and frequency of replies to statements about pain information Item Mean SD % of repliesa 1–2 Source of information Information about pain should be provided by the medical oncologist Information about pain is routinely provided by the medical oncologist Information about pain should be provided by the nurse Information about pain is routinely provided by the nurse Manner of providing information Information about pain should be provided directly to the patient Information about pain is routinely provided directly to the patient Information about pain should be provided to the patient through family members Information about pain is routinely provided to the patient through family members Oncologists should confirm that the patient has understood the information provided about pain Oncologists actively confirm that their patients have understood the information provided about pain It would be desirable to provide patients with written information about their disease, cancer treatments and the side effects of medication Patients are routinely provided with written as well as verbal information about their disease It would be desirable to provide patients with written information about their pain, their treatment and the side effects of pain medication Patients are routinely provided with written as well as verbal information about pain 3 4–5 4.5 0.5 0 0 100 4.4 0.5 0 0 100 2.2 0.8 68 23 9 1.9 0.7 82 18 0 4.9 0.3 0 0 100 4.1 0.7 5 9 86 1.4 0.7 91 9 0 2.3 0.8 59 36 5 4.6 0.5 0 0 100 2.2 0.8 68 27 5 4.5 0.9 5 5 91 2.7 1.0 50 23 27 4.5 0.9 5 5 91 1.7 0.7 95 0 5 a Items were rated on a 5-point Likert scale, where 1 = strongly disagree and 5 = strongly agree. that oncologists should confirm that patients have understood the information provided, and that they should give their patients written information about their cancer and pain treatments, as well as the side effects of these treatments. However, they strongly disagreed with the statement that oncologists do actively confirm that patients understand what they are told, or that oncologists provide any sort of written in addition to verbal information. In the area of pain evaluation (Table 2), all of the panelists agreed that the nurse’s opinion should be and is taken into account when evaluating pain in hospitalized patients. They also generally agreed that both patients’ preferences and the opinion of informal care givers should be considered in this context, however a considerable proportion (50% and 41%, respectively) did not agree that this was generally the case. There was total agreement that, ideally, consideration should be given to other aspects that may be involved in the perception of pain, such as anxiety, depression and social support, and over two-thirds of the group thought this was routinely done. The oncologists also strongly agreed that all sites and diagnoses of a patient’s pain should be evaluated, and that pain should be evaluated at each office visit, however only 50% and 41%, respectively, agreed that that these evaluations were performed in routine clinical practice. With regard to how pain should be measured, all agreed with the use of the Visual Analog Scale (VAS) in pain evaluation, but over 40% either disagreed or had no opinion as to whether this instrument is routinely used in clinical practice. There was wide dispersion of opinion about the desirability of using validated instruments other than the VAS in pain evaluation, and most of the group thought such instruments are not routinely used. With regard to issues related with pain management (Table 3) the panelists’ perceptions of ‘‘real’’ and ‘‘ideal’’ practice generally coincided with regard to how opioids should be chosen: based on the scientific evidence, their efficacy and speed of action. They mostly agreed that opioids should be chosen based on their side effects and regardless of the patient’s age, but one half did not agree with the statement that side effects are, in fact, routinely considered in opioid prescription, and a slightly larger proportion did not disagree with the statement that opioids are routinely chosen based on age. Although all agreed that the need for a prescription for narcotics should not influence the choice of opioid analgesia, a substantial minority (36%) were undecided as to whether this was really the case. There was little disagreement, however, that oncologists should and do write prescriptions for pain medication during an office visit. There was total consensus that effective analgesia requires administration at fixed intervals rather than ‘‘on demand’’, and that patients are routinely treated accordingly. Panelists mostly agreed that cancer pain 355 J. Carulla Torrent et al. / European Journal of Pain 11 (2007) 352–359 Table 2 Mean ratings and frequency of replies to statements about pain evaluation Item Mean Aspects to consider in pain evaluation The nurse’s opinion should be taken 4.7 into account when estimating pain in hospitalised patients 4.1 The nurse’s opinion is taken into account when estimating pain in hospitalised patients The patient’s preferences should be taken into account when modifying the plan for pain medication The patient’s preferences are routinely taken into account when treating pain The opinion of informal caregivers (family members,. . .) should be considered in pain evaluation The opinion of informal caregivers is routinely considered in pain evaluation When evaluating pain consideration should be given to other aspects (anxiety, depression, social support, personality) that may be implicated in the patient’s perception of pain When evaluating pain consideration is given to other aspects (anxiety, depression, social support, personality) that may be implicated in the patient’s perception of pain Frequency and sites of evaluation All sites and diagnoses of a patient’s pain should be evaluated All sites and diagnoses of a patient’s pain are routinely evaluated The patient’s pain should be systematically evaluated at each visit Pain evaluation is a routine practice in medical oncologists’ offices Use of instruments to evaluate pain The VAS should be used in pain evaluation The VAS is routinely used in evaluating a patient’s pain Validated scales or instruments other than the VAS should be used in pain evaluation Validated scales or instruments other than the VAS are routinely used in pain evaluation 4.5 SD % of replies a 1–2 4–5 3 Item 0.4 0 0 100 0.7 5 9 86 0.6 0 5 95 3.5 0.7 9 41 50 4.3 0.7 0 14 86 3.6 4.8 3.6 0.8 0.4 1.0 9 0 18 32 0 14 59 100 68 4.9 0.3 0 0 100 3.5 0.7 9 41 50 4.8 0.4 0 0 100 3.2 0.8 18 41 Table 3 Mean ratings and frequency of replies to statements about pain management 41 4.9 0.3 0 0 100 3.7 0.8 5 36 59 3.0 0.9 27 50 23 1.7 0.8 81 19 0 a Items were rated on a 5-point Likert scale, where 1 = strongly disagree and 5 = strongly agree. Mean SD % of repliesa 1–2 Choice and use of analgesia Opioids should be chosen based on the scientific evidence about their use Opioids are routinely chosen based on the scientific evidence Opioids should be chosen based on their efficacy Guidelines for analgesics that have been shown to have the greatest efficacy are not used Opioids should be chosen based on their speed of action Opioids are routinely chosen based on their speed of action Opioids should be chosen based on their side effects Opioids are routinely chosen based on their side effects The choice of opioids should be made regardless of the patient’s age Opioids are routinely chosen based on the patient’s age Opioids should be chosen regardless of whether or not a prescription for narcotics is required Opioids that do not require a prescription for narcotics are routinely chosen Official prescriptions for analgesics should be written by the oncologist during an office visit Official prescriptions for analgesics are written by the oncologist during an office visit It is important to maintain a constant level of analgesia for it to be effective Analgesia is generally prescribed at fixed intervals (e.g., every 6 hours), which allows good pain control Patients are generally prescribed analgesia ‘‘on demand’’ or ‘‘in case of pain’’ Cancer pain should be treated in accordance with the guidelines of the WHO ladder Cancer pain is treated in accordance with the guidelines of the WHO ladder In general, cancer pain is undertreated Use of other medications/treatments Pain control can be improved by using other drugs (e.g., anti-anxiety agents, sedatives, antidepressants, neuroleptics) in addition to analgesics All patients should receive medications (e.g., anti-anxiety agents, sedatives, antidepressants, neuroleptics) in addition to analgesics for pain control 3 4–5 4.9 0.3 0 0 100 4.2 0.6 0 14 86 4.9 0.3 0 0 100 2.2 1.0 67 19 14 4.4 0.8 5 5 91 4.0 0.6 5 0 95 4.0 0.8 5 18 77 3.4 0.9 14 36 50 4.2 0.8 5 14 82 2.5 0.9 45 41 14 5.0 0.2 0 0 100 2.1 0.8 64 36 0 4.4 0.8 5 10 85 3.9 0.7 0 29 71 4.9 0.3 0 0 100 4.3 0.4 0 0 100 1.7 0.8 82 18 0 3.9 1.2 19 0 81 3.7 0.6 5 24 71 3.7 1.0 9 41 50 4.9 0.3 0 0 100 3.8 1.2 18 9 73 (continued on next page) 356 J. Carulla Torrent et al. / European Journal of Pain 11 (2007) 352–359 Table 3 (continued) Item Mean SD % of repliesa 1–2 3 4–5 Psychotropic drugs are prescribed less often than necessary Psychotropic drugs are prescribed more often than necessary Other treatments for cancer pain (such as relaxation techniques, cognitive restructuring, rehabilitation) should be used in addition to medical treatment of pain In addition to medical treatment, other treatments (such as relaxation techniques, cognitive restructuring, rehabilitation) are frequently used for cancer pain Laxatives should routinely be prescribed in association with opioid treatment Laxatives are routinely prescribed in association with opioid treatment 3.1 0.8 18 50 32 2.8 1.0 45 27 27 4.2 0.7 0 18 82 1.8 0.6 86 14 0 4.9 0.3 0 0 100 3.6 0.9 14 24 62 4.4 1.1 9 9 82 4.1 0.9 9 9 82 4.1 1.1 9 14 77 3.5 1.2 14 23 64 2.7 1.2 52 14 33 4.6 0.5 0 0 100 3.5 0.8 18 23 59 4.9 0.3 0 0 100 3.6 0.8 14 14 73 4.5 0.5 0 0 100 3.4 0.9 29 10 62 Referral to other services Oncologists should refer their patients, when necessary, to palliative care units Oncologists refer their patients to palliative care units when necessary Oncologists should refer their patients to pain units when necessary Oncologists refer their patients to pain units whey they think it is necessary Medical oncologists frequently refer their patients to pain units inappropriately Oncologists should refer their patients to psychosocial support services when necessary Oncologists refer their patients to psychosocial support services when necessary Other The patient with uncontrolled pain should be able to visit the oncologist on demand Patients with uncontrolled pain can visit the oncologist on demand Specific training in cancer pain would help health workers to manage pain well Oncology residents receive specific training in pain management a Items were rated on a 5-point Likert scale, where 1 = strongly disagree and 5 = strongly agree. should be (81%) and is (71%) treated in accordance with the guidelines of the WHO ladder (17). At the same time, however, 50% of the panelists agreed that cancer pain is generally undertreated, and very few disagreed with this statement. The use of other medications and treatments, in addition to opioids, was the subject of some discrepancy. There was complete agreement among panelists that pain control can be improved by using psychotropic drugs (anti-anxiety agents, sedatives, antidepressants and neuroleptics), as adjuvant treatment, and 73% believed that all patients should receive these types of medications. However, panelists differed in their perceptions of the real situation, with 32% agreeing that psychotropic drugs are underprescribed and 27% saying they are overprescribed. A more striking discrepancy between what is considered real and ideal practice was seen with regard to the use of non-pharmacological treatments for cancer pain. Whereas over 80% agreed that it is desirable to use such treatments as relaxation techniques, cognitive restructuring and rehabilitation, 86% disagreed that this is frequent in practice. Another area, where real and ideal practice diverge to some extent refers to the use of laxatives: all panelists believe they should routinely be prescribed in association with opioid treatment, but 4 out of 10 did not agree that this constitutes routine clinical practice. In the area of referral to other services, it was generally thought that oncologists should and do refer their patients to palliative care units when necessary, but the situation was less clear with regard to referral to pain units or psychosocial support services. Although over three-quarters of the group agreed that oncologists should refer patients to pain units when necessary and almost two-thirds thought they actually did so, about one-third thought medical oncologists frequently make inappropriate referrals to pain units. In the case of psychosocial support services, everyone agreed that patients should be referred when necessary, but only 59% said that oncologists generally do this. There was universal agreement that patients with uncontrolled pain should be able to visit the oncologist ‘‘on demand’’, and about three-quarters of the panelists thought this was usually the case in practice. Finally, this group of oncologists strongly agreed that health workers would benefit from training in cancer pain management, but only 62% considered that oncology residents actually receive specific training in this area. 4. Discussion The results of this survey of the perceptions of a group of experienced medical oncologists in Spain suggest a number of areas where routine clinical practice falls short of the ideal. In the area of information, it was seen that few panelists thought that patients routinely receive written information regarding pain or their disease, or that oncologists make any attempt to actively confirm that their patients have understood what they are told. Studies have shown that patients often fail to report pain, that they may believe pain is ‘‘inevitable’’ and should be tolerated, and that non-compliance with J. Carulla Torrent et al. / European Journal of Pain 11 (2007) 352–359 treatment recommendations may occur due to forgetfulness, fears of opioid addiction or tolerance, or concerns about side effects (Ward et al., 1993; Thomason et al., 1998). Clearly, optimal pain control cannot be achieved without an accurate report from the patient. Thus, it is important that physicians educate their patients about pain control, actively elicit information about pain intensity and adherence to the treatment plan, and encourage them to ask questions. Given the limited duration of a typical specialist visit, the provision of written information that the patient can consult at home could be especially helpful. Despite the panelists’ strong support for systematic pain evaluation, only 40% believed this constitutes routine clinical practice. A slight majority of panelists agreed that the VAS is generally used, and no one thought that any other instruments for pain evaluation are routinely employed. A number of studies have shown that, in other countries as well, failure to systematically evaluate pain is one of the most important barriers to pain management (MacDonald et al., 2002; Von Roenn et al., 1993). A study in a US outpatient oncology practice found that a simple intervention – training health assistants to measure and document pain using the VAS – increased pain score documentation from 0% to over 70% (Rhodes et al., 2001). It was also believed that patient preferences are frequently not taken into account when modifying the plan for pain medication and that factors such as anxiety, depression and social support are not necessarily considered when evaluating a patient’s pain. Little information is available on patient participation in planning treatment for cancer pain, although a small qualitative study in Sweden suggests that palliative care patients would welcome taking an active part in planning both pharmacological and non-pharmacological treatments (Boström et al., 2004). In the area of pain management, most panelists agreed with statements supporting the choice of opioids based on the scientific evidence, but more than half either agreed with or were neutral about the statement that opioids are routinely chosen based on the patient’s age. This suggests that some oncologists may have inappropriate concerns about administering strong opioids to elderly patients, for fear of hastening their death, or to younger patients, for fear of possible addiction. Just half of the panel agreed that opioids are routinely chosen based on their side effects, although over three-quarters thought they should be. Opioids are associated with a number of adverse side effects, constipation being one of the most common. That oncologists need to pay more attention to this problem is suggested by the fact that 4 out of 10 panelists did not agree that laxatives are routinely prescribed in association with opioid treatment. Most of the panel agreed that cancer pain should be and is, in fact, treated in accordance with the guidelines 357 of the WHO ladder. Nevertheless, fully half of the panelists agreed with the statement that cancer pain is undertreated in Spain. Studies in the US (Cleeland et al., 1994; Von Roenn et al., 1993), Italy (Di Maio et al., 2004) and France (Larue et al., 1995) indicate that this is a common situation. In Spain, a review of clinical records of terminal cancer patients receiving care in an urban health setting showed that 76% and 50% of patients had received infra-therapeutic doses of codeine and morphine, respectively; most of these patients had been attended in Primary Care and received their prescriptions for pain medication from general practitioners (Rispau Falgas et al., 1999). Many studies have identified obstacles to adequate pain treatment, including lack of systematic pain assessment, regulatory restrictions on opioid use, physician and patient concerns about side effects or addiction, patient reluctance to report pain or to take opioids, and patient noncompliance or misunderstanding of instructions for analgesics (MacDonald et al., 2002; Vainio, 1995; Von Roenn et al., 1993). Further investigation is needed to explore which obstacles are most important in the Spanish health system and how they may be overcome. Referral to pain units and psychological support services is another area where there is room for improvement. In the case of pain units, the discrepancy between real and ideal may be due to the fact that not all Spanish hospitals have such units, and even those that do may lack clear criteria for patient referral. The latter hypothesis may explain why one third of the panelists consider that patients are frequently referred inappropriately to pain units. Less than 10% of cancer patients will require invasive treatments for pain control although, for those that do (difficult cases involving neuropathic pain, for example), it is important to institute early treatment, without waiting for the failure of noninvasive treatments. In the case of psychological support services, the failure to refer may be related with the physician’s lack of knowledge of the patient’s personal situation or to the absence of such programs in some hospitals. Non-pharmacological treatments (NPTs) for cancer pain are also clearly underused, in the opinion of the panel. Eight out of 10 respondents believed treatments such as relaxation techniques and cognitive restructuring should be used to treat cancer pain, but no one agreed that these treatments are frequently used. While studies have suggested that NPTs such as patient psychoeducation, supportive psychotherapy and cognitivebehavioral interventions can be effective for managing pain and improving the quality of life in oncology populations (Fawzy, 1999; Meyer and Mark, 1995; Thomas and Weiss, 2000), it appears that health-care professionals rarely recommend such techniques (Zaza et al., 1999). In Spain, although some large hospitals have ‘‘psycho-oncology’’ units which incorporate some of 358 J. Carulla Torrent et al. / European Journal of Pain 11 (2007) 352–359 these techniques, they are generally reserved for problematic patients with psychiatric pathologies, and rarely carry out therapeutic interventions to enable patients to cope better with their pain. Another reason for underreferral or inappropriate referral to services outside the oncology unit may be poor physician–patient communication, either due to lack of time or because physicians are primarily interested in obtaining information needed to make treatment decisions, but not necessarily for symptom management (Rogers and Todd, 2002). Specific training in cancer pain was considered useful by all the oncologists surveyed, yet only 62% said that oncology residents receive such training. Even assuming this to be the case, it is not clear if the training received is adequate, as the questionnaire contained no item regarding the quality of training. Studies in other countries have shown that pain education in medical school or residency programs is often considered poor or mediocre (MacDonald et al., 2002; Von Roenn et al., 1993), thus this may also represent an important area for action. This study clearly has certain limitations. The panel of experts was not randomly chosen and thus cannot be considered a representative sample of Spanish medical oncologists. In addition, our inferences about what constitutes ‘‘real’’ clinical practice are based on panelists’ opinions, which may only reflect their experiences in their own hospital. Nevertheless, all of the panelists had extensive clinical experience treating cancer patients, and had been chosen to participate in the ALGOS group by virtue of their prominence in the field of Spanish oncology. The 22 hospitals represented in this survey see an average of 800 cancer patients annually, for a total of 17,600 patients, or 20% of all cases of cancer produced in Spain per year. Their opinions most likely do represent their experiences in their own settings, however the fact that they were almost all from different hospitals with wide geographic distribution means that an extensive diversity of opinions is represented. Furthermore, these oncologists’ perception of what constitutes routine clinical practice probably represents a ‘‘best case’’ scenario since they are almost all from large university hospitals. If physicians from smaller hospitals had been included, the gap between ‘‘real’’ and ‘‘ideal’’ would likely be even wider. This Delphi study of Spanish oncologists’ perceptions of how cancer pain is and should be managed suggests that, even in large university hospital settings, there may be areas, where ‘‘real’’ clinical practice lags considerably behind the ‘‘ideal.’’ Most notably, medical oncologists need to improve communication with patients, ensuring that they receive both written and verbal information about their disease and the plan for pain treatment; physicians also need to elicit patients’ concerns about pain medication so that common barriers to compliance with treatment can be addressed. Systematic evaluation of pain at each visit, using the VAS or other validated instruments, is another area that needs to be further emphasized. Although the evidence base for the use of NPTs for pain control is much more limited than in the case of analgesics, the panelists strongly agreed that these techniques are potentially beneficial, but are rarely used for cancer pain. Given that such therapies are generally safer and less expensive than pharmacological treatment, oncologists should familiarize themselves with the resources available for the use of these techniques for pain control. Incorporating patient preferences into the pain management plan is another challenging area that should not be ignored. Finally, both medical students and residents could benefit from specific training in the evaluation and treatment of cancer pain. Acknowledgements Financial support for the Delphi study was received from Janssen-Cilag. The authors also thank Pablo Lázaro y de Mercado for constructive suggestions on the manuscript, as well as the following members of the ALGOS group for their participation in this project: Albert Abad, Antonio Antón, Enrique Aranda, Alfredo Carrato, Manuel Constenla, Juan Jesús Cruz, Francisco Javier Dorta, Manuel González-Barón, Vicente Guillén, José Manuel López, Bartomeu Massuti, José Andrés Moreno, Adolfo Murias, Gumersindo Pérez and Alejandro Tres. References Addington-Hall J, McCarthy M. Dying from cancer: results of a national population-based investigation. Palliat Med 1995;9:295–305. Ahmedzai S. New approaches to pain control in patients with cancer. Eur J Cancer 1997(Suppl 6):S8–S14. Bassols A, Bosch F, Campillo M, Canellas M, Banos JE. An epidemiological comparison of pain complaints in the general population of Catalonia (Spain). Pain 1999;83:9–16. Benedetti C, Brock C, Cleeland C, Coyle N, Dube JE, Ferrell B, et al. National comprehensive cancer network. NCCN practice guidelines for cancer pain. Oncology (Williston Park) 2000;14:135–50. Boström B, Sandh M, Lundbert D, Fridlund B. Cancer-related pain in palliative care: patients’ perceptions of pain management. J Adv Nurs 2004;45:410–9. Canellas M, Bosch F, Bassols A, Rue M, Banos JE. The prevalence of pain in hospitalized patients. Med Clin (Barc) 1993;101:51–4 [article in Spanish]. Carr DB, Goudas LC, Balk EM, Bloch R, Ioannidis JPA, Lau J. Evidence report on the treatment of pain in cancer patients. J Natl Cancer Inst Monogr 2004;32:23–31. Carr DB. The development of national guidelines for pain control: synopsis and commentary. Eur J Pain 2001;5(Suppl A):91–8. Catala E, Reig E, Artes M, Aliaga L, López JS, Segu JL. Prevalence of pain in the Spanish population: telephone survey in 5000 homes. Eur J Pain 2002;6:133–40. J. Carulla Torrent et al. / European Journal of Pain 11 (2007) 352–359 Cleeland CS, Gonin R, Hatfield AK, Edmonson JH, Blum RH, Stewart JA, et al. Pain and its treatment in outpatients with metastatic cancer. New Engl J Med 1994;330:592–6. Di Maio M, Gridelli C, Gallo C, Manzione L, Brancaccio L, Barbera S, et al. Prevalence and management of pain in Italian patients with advanced non-small-cell lung cancer. Br J Cancer 2004;90:2288–96. Farre J, Badenas JM, Lamas X, Ugena B, Cami J. Treatment of acute pain in a general hospital: opinions of physicians and nurses. Rev Clin Esp 1992;190:152–6 [article in Spanish]. Fawzy FI. Psychosocial interventions for patients with cancer: what works and what doesn’t. Eur J Cancer 1999;35:1559–64. Green CR, Wheeler JRC, LaPorte F. Clinical decision making in pain management: Contributions of physician and patient characteristics to variations in practice. J Pain 2003;4:29–39. Jacox A, Carr DB, Payne R. New clinical-practice guidelines for the management of pain in patients with cancer. New Engl J Med 1994;330:651–5. Larue F, Colleau SM, Brasseur L, Cleeland CS. Mulicentre study of cancer pain and its treatment in France. Br Med J 1995;310:1034–7. Levy MH. Pharmacologic treatment of cancer pain. New Engl J Med 1996;335:1124–32. López de Maturana A, Morago V, San Emeterio E, Gorostiza J, Arrate AO. Attitudes of general practitioners in Bizkaia, Spain, towards the terminally ill patient. Palliat Med 1993;7:39–45. MacDonald N, Ayoub J, Farley J, Foucault C, Lesage P, Mayo NA. Quebec survey of issues in cancer pain management. J Pain Symptom Manage 2002;23:39–47. Meyer TJ, Mark MM. Effects of psychosocial interventions with adult cancer patients: a meta-analysis of randomized experiments. Health Psychol 1995;14:101–8. Rhodes DF, Koshy RC, Waterfield WC, Wu AW, Grossman SA. Feasibility of quantitative pain assessment in outpatient oncology practice. J Clin Oncol 2001;19:501–8. 359 Rispau Falgas A, Soler Vila M, Garcia Bayo I, Valles Plasencia A, Saladich Puig R, Martı́nez Alemany V. How we use opioid drugs on patients with neoplasms. Aten Primaria 1999;24:537–9 [article in Spanish]. Rogers M, Todd C. Information exchange in oncology outpatient clinics: source, valence and uncertainty. Psycho-Oncology 2002;11:336–45. Schug SA, Zech K, Dorr U. Cancer pain management according to WHO analgesic guidelines. J Pain Symptom Manage 1990;5:27–32. Thomas EM, Weiss SM. Nonpharmacological interventions with chronic cancer pain in adults. Cancer Control 2000;7:157–64. Thomason TE, McCune JS, Bernard SA, Winer EP, Tremont S, Linkley CM. Cancer pain survey: patient-centered issues in control. J Pain Symptom Manage 1998;15:275–84. Vainio A. Treatment of terminal cancer pain in France: a questionnaire study. Pain 1995;62:155–62. Valero J, Palacios P, Vázquez-Barro A, López-Suso E, Carpintero D. Study of prevalence of pain at the Juan Canalejo Hospital in La Coruña. An Med Interna 1995;12:584–8 [article in Spanish]. Von Roenn JH, Cleeland CS, Gonin R, Hatfield AK, Pandya KJ. Physician attitudes and practice in cancer pain management: a survey from the Eastern Cooperative Oncology Group. Ann Int Med 1993;119:121–6. Ward SE, Goldberg N, Miller-McCauley V, Mueller C, Nolan A, Pawlik-Plank D. Patient-related barriers to management of cancer pain. Pain 1993;52:319–24. Yun YH, Heo DS, Lee IG, Jeong HS, Kim HJ, Kim SY. Multicenter study of pain and its management in patients with advanced cancer in Korea. J Pain Symptom Manage 2003;25:430–7. Zaza C, Sellick SM, Willan A, Reyno L, Browman GP. Health care professionals’ familiarity with non-pharmacological strategies for managing cancer pain. Psycho-Oncolgy 1999;8:99–111. Zenz M, Zenz T, Tryba M, Strumpf J. Severe undertreatment of cancer pain in Germany – a three years’ survey. J Pain Symptom Manage 1995;10:187–91.