- Email: [email protected]
Impact of setting of care on pain management in patients with cancer: a multicentre cross-sectional study
original article
Annals of Oncology 21: 2088–2093, 2010 doi:10.1093/annonc/mdq155 Published online 31 March 2010
Impact of setting of care on pain management in patients with cancer: a multicentre cross-sectional study D. Sichetti1 , E. Bandieri2 , M. Romero1 , K. Di Biagio1, M. Luppi3*, M. Belfiglio1, G. Tognoni1 & C. I. Ripamonti4* for ECAD Working Group 1
Department of Clinical Pharmacology and Epidemiology, Consorzio Mario Negri Sud, Santa Maria Imbaro (Chieti); 2Palliative Care Unit, Ausl Modena, Modena; Department of Oncology, Haematology and Respiratory Diseases, Azienda Ospedaliera Universitaria, University of Modena and Reggio Emilia, Modena; 4Supportive Care in Cancer Unit, IRCCS Foundation, National Cancer Institute, Milano, Italy 3
Background: No study has so far addressed whether differences do exist in the management of cancer-related pain in patients admitted to oncology and non-oncology settings. Patients and methods: A multicentre cross-sectional study in 48 Italian hospitals has enrolled 819 patients receiving analgesic therapy for cancer-related pain. Demographics and clinical and analgesic therapy information have been prospectively collected by standardized forms. Adequacy of pain management has been evaluated by the Pain Management Index (PMI). Results: Differences in the analgesic drug administration according to settings of care have been evident, non-opioids more frequently being administered in non-oncology units (19.6% versus 7.0%; P < 0.0001), while strong opioids are more frequently used in the oncology units (69.5% versus 51.9%; P < 0.0001). The number of patients receiving inadequate therapy (PMI < 0) has lowered in oncology compared with non-oncology units (11.3% versus 18.8%; P = 0.0024). Results of multiple logistic regression analysis have shown that the admission to non-oncology setting [odds ratio (OR) = 1.75, 95% confidence interval (CI) = 1.15–2.67; P = 0.0096] and the absence of metastatic disease (OR = 1.60, 95% CI = 1.04–2.44; P = 0.0317) were independent factors associated with an increased risk of receiving an inadequate analgesic therapy. Conclusion: Oncology wards provide the most adequate standard of analgesic therapy for cancer-related pain. Key words: analgesic drugs, neoplasms, pain, settings of care
introduction Pain is one of the most feared and burdensome symptoms in cancer patients and unrelieved pain continues to be a substantial worldwide public health concern. Consistent with this, a systematic review of the literature investigated the prevalence of pain in different disease stages and types of cancer during the period 1966–2005 [1]. The pooled data from 52 articles showed that pain prevalence was 64% in patients with metastatic, advanced or terminal phase disease, 59% in patients on anticancer treatment and 33% in patients after curative treatment. No difference in pain prevalence was found *Correspondence to: Prof. M. Luppi, Department of Oncology, Hematology and Respiratory Diseases, Azienda Ospedaliera Universitaria, University of Modena and Reggio Emilia, Modena, Italy. Tel: +39-059-4225570; Fax: +39-059-4224549; E-mail: [email protected] or Dr. C. I. Ripamonti, Supportive Care in Cancer Unit, IRCCS Foundation, National Cancer Institute, Via Venezian n 1, 20133 Milano, Italy. Tel: +39-02-2390-3640; Fax: +39-02-2390-3656; E-mail: [email protected]
These authors contributed equally.
between the patients during anticancer treatment and those in advanced or terminal phase of the disease. In the 18 studies reporting the pain severity, one-third of the patients rated their pain as moderate to severe [1]. Moreover, in the systematic review of the literature carried out by Deandrea et al. [2] on studies published from 1994 to 2007, nearly half of the cancer patients were under-treated with a high variability across study designs and clinical settings. Recent studies conducted both in Italy and in Europe [3, 4] confirmed these data, showing that pain was present in all phases of cancer disease (early and metastatic) and was not adequately treated in a significant percentage of patients, ranging from 56% to 82.3%. In particular, Apolone et al. [5] evaluated prospectively the adequacy of analgesic care of cancer patients by means of Pain Management Index (PMI) [6] in 1802 valid cases of in- and outpatients with advanced/metastatic solid tumour enrolled in 110 centres specifically devoted to cancer and/or pain management (oncology/pain/palliative centres or hospices). The study showed that patients were still classified as potentially under-treated in 25.3% of the cases (range 9.8%–55.3%).
ª The Author 2010. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected]
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 19, 2015
original article
Received 1 March 2010; accepted 4 March 2010
original article
Annals of Oncology
patients and methods study design and setting A multicentre cross-sectional study has been carried out in 48 hospitals, located in 13 different Italian regions, to identify patients receiving analgesics. In each hospital, a multidisciplinary working group, made up of physicians, nurses and pharmacists, has been established. Each local working group, promoting this survey as a hospital pharmaco-surveillance activity, has predetermined index days to carry out the survey and coordinated participant wards. Overall, in the 164 clinical wards involved, all patients treated with analgesic therapy, irrespective of age and types of pain, have been considered eligible for this project. The patients, treated with an analgesic drug according to the usual practice, have been sampled in six index days (from 12 February 2007 through 29 July 2008). Wards have been divided into two area-type clusters, namely oncology and non-oncology wards, the latter including surgery, internal medicine and orthopaedics wards. This project has been authorized by the Hospital General Management and by the Local Ethical Committee. For each patient included in the study, data have been obtained according to Italian law about privacy (D.Lgs. 196/2003), so that all patients have been recruited anonymously to the central data analysis.
data collection Information regarding patients and wards have been collected using standardized forms. For each patient involved in the survey, demographics (age and gender), clinical treatment (type of pain, primary cancer and presence of metastasis) and analgesic treatment (type of analgesic drug prescribed and administered in the last 24 h, route of administration, dosage and adjuvant treatment modality) have been obtained from medical records. Patients have been interviewed by a pharmacist—considered an independent health professional not involved in patients’ care—concerning their perception of pain control and their pain intensity at interview and the worst intensity of pain perceived during the previous 24 h. Pain intensity has been estimated by using a four-level Verbal Rating Scale as no pain,
Volume 21 | No. 10 | October 2010
mild pain, moderate pain and severe pain. We have defined and scored pain levels as follows—0 points: no pain; 1 point: mild pain; 2 points: moderate pain and 3 points: severe pain. Health care professionals (nurses and/or clinicians) have been asked about their perception on the effectiveness of the analgesic treatment administered and the possible reasons for the absence of treatment response. Moreover, patients have also been asked about their relationship with health care professionals, and, specifically, whether they have communicated their pain, whether they have received a treatment after communication and the longest time interval they have had to wait before the initiation of treatment. Finally, patients have also evaluated the efficacy of the analgesic therapy received. Reasons for failed interviews (refusal, patient’s absence, cognitive limitations/impairment, language barrier or other reasons) have been collected by the interviewer.
pain management index To evaluate the adequacy of pain management according to the World Health Organization (WHO) guidelines [11], PMI has been assessed [6]. The Cleeland’s PMI compares the analgesic drug with the level of pain reported by the patients and is computed by subtracting pain scores from analgesic scores [6]. The analgesics were scored as follows: 0 points = no analgesic; 1 point = WHO I step (non-opioid); 2 points = WHO II step (weak opioids, such as codeine and tramadol) and 3 points = WHO III step (strong opioids, such as morphine, oxycodone, methadone, fentanyl, hydromorphone and buprenorphine). We have defined and scored pain levels as follows: 0 points = no pain; 1 point = mild pain; 2 points = moderate pain and 3 points = severe pain. PMI ranges from 23 (patients with severe pain receiving no drug at all) to +3 (patients receiving strong opioids and reporting pain relief). Negative scores indicate inadequate orders for analgesic drugs and have been considered as under-treatment, whereas scores of 0 points and higher have been considered as indicators of acceptable treatment.
statistical analysis Patients’ baseline characteristics (gender, age, primary cancer and presence of metastasis), analgesic treatment (non-opioids, weak opioids and strong opioids) and (in)adequacy of cancer pain management (computed using PMI) have been reported as percentages in the two different settings of care (oncology and non-oncology wards) and have been compared with chi-square test. For univariate analysis, the following variables were considered: gender, age, primary cancer, metastasis and setting of care. In order to identify independent factors associated with inadequate pain treatment (PMI < 0), a multiple logistic regression model has been used to control for the same variables tested in the univariate analysis, simultaneously. Multivariate models were also constructed using stepwise selection methods. The results have been expressed as odds ratios (ORs) and 95% confidence intervals (95% CIs). P values <0.05 have been considered significant. All analyses have been carried out using SAS Statistical Package Release 9.2 (SAS Institute, Cary, NC).
results A total of 819 patients were included in the analysis. Of them, 55.4% were older than 65 years and 51.6% of them were male. The characteristics of the patients, 426 in oncology and 393 in non-oncology units, grouped by setting of care are reported in Table 1. The two patient groups considered have shown statistically significant differences (P < 0.001) in relation to
doi:10.1093/annonc/mdq155 | 2089
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 19, 2015
Moreover, different studies [7–9] have clearly shown a lack of adequate pain management in the majority of the Italian hospitals and have reinforced the need for the implementation of organizational and educational interventions, in order to emend this widespread malpractice. However, no study has so far been carried out in Italy with the aim to compare the modalities of treatment of cancerrelated pain in patients admitted to different speciality settings of care. The ECAD-O (Epidemiologia Clinico-Assistenziale del Dolore in Ospedale) (Clinical-Caring Epidemiology of Pain in Hospital) group has been established to activate a surveillance on pain management, as part of a routine care, by creating a multicentre network of hospitals and a multidisciplinary working group. Specifically, a survey has been carried out in order to investigate the characteristics of inpatients treated with analgesic drugs and to evaluate the perception of the level of pain control both from patients and health care professionals [10]. The preliminary results of this ECAD-O survey [10] have shown that cancer pain remains one of the main causes of analgesic therapy administration both in oncology and in non-oncology (surgery, internal medicine and orthopaedics) wards. The aim of the present research study was to determine the impact of setting of care on pain management in patients with cancer.
original article
Annals of Oncology
certain clinical features: age, primary cancer and metastatic disease (Table 1). The types of analgesic therapy, including non-opioid, weak opioid and strong opioid drugs administered to the patients in the two different settings of care, have also been analysed and differences between the two settings have are evident. In particular, as reported in Figure 1, the difference was statistically significant for non-opioid drugs that have been administered more frequently in non-oncology than in oncology units (19.6% versus 7.0%; P < 0.0001) and for the use of strong opioids that has resulted more frequent in the oncology setting (69.5% versus 51.9%; P < 0.0001). The frequencies of different types of analgesic therapy, administered to the two patient groups, following the evaluation of the intensity of pain recorded in the last 24 h
Table 1. Characteristics of the patients in the two settings of care
Oncology units (n = 426), n (%)
P value
180 (45.8) 213 (54.2)
216 (50.7) 210 (49.3)
151 101 107 34
214 141 62 9
0.1607
<0.0001 (38.4) (25.7) (27.2) (8.7)
(50.2) (33.1) (14.6) (2.1) <0.0001
19 (4.9) 72 (18.4) 135 (34.5) 60 (15.4) 30 (7.7) 30 (7.7) 45 (11.5) 2
30 (7.1) 79 (18.6) 90 (21.2) 60 (14.2) 22 (5.2) 52 (12.3) 91 (21.5) 2
146 (37.9) 239 (62.1) 8
98 (23.1) 326 (76.9) 2
<0.0001
Table 2. Distribution of the patients according to the analgesic treatment received and pain intensity in the two settings of care Pain 24 h
Non-oncology units None Mild Moderate Severe Total Oncology units None Mild Moderate Severe Total
Analgesic treatment received Strong Weak Nonopioids, opioids, opioids, n n n
Total
12 18 21a 26a 77
28 25 32 27a 112
22 46 67 69 204
62 89 120 122 393
7 8 9a 6a 30
14 25 28 33a 100
51 72 68 105 296
72 105 105 144 426
a
Patients with inadequate treatment according to PMI. PMI, Pain Management Index.
Table 3. Distribution of patients according to the adequacy/inadequacy of the analgesic treatment received and the pain intensity as reported by the patients (PMI) in the two settings of care
PMI <0 (inadequate) ‡0 (adequate) Figure 1. Analgesic treatment received in the two settings of care.
2090 | Sichetti et al.
Non-oncology units (n =393), n (%)
Oncology units (n =426), n (%)
74 (18.8) 319 (81.2)
48 (11.3) 378 (88.7)
P value
0.0024
PMI, Pain Management Index.
Volume 21 | No. 10 | October 2010
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 19, 2015
Gender Female Male Age £64 65—74 75—84 ‡85 Primary cancer Haematological Lung Gastrointestinal Genito-urinary Bone and soft tissues Breast Others Missing Metastasis No Yes Missing
Setting of care Non-oncology units (n = 393), n (%)
before the interview, have also been compared. As reported in Table 2, strong opioids have been used in 69 of 122 (56.5%) patients with severe pain in non-oncology units and in 105 of 144 (72.9%) patients with pain of similar intensity in oncology units. Non-opioids have been used in 26 patients (21.3%) with severe pain in non-oncology units and in only 6 (4.2%) patients in oncology units. Moreover, strong opioids have also been used for the treatment of mild pain in 46 (51.7%) and 72 (68.6%) patients in non-oncology and oncology units, respectively. By applying the PMI criteria to our study population, the number of patients receiving an appropriate treatment (PMI ‡ 0) has increased in oncology setting (88.7%) than in nononcology setting (81.2%). On the contrary, inadequate therapy (PMI < 0) has been administered to a lower number of patients in oncology units (11.3%) compared with non-oncology units (18.8%). These differences were statistically significant (P = 0.0024; Table 3). Univariate analysis showed that the admission to nononcology units and the absence of metastatic disease showed a significant relationship with respect to receiving an inadequate analgesic treatment (Table 4). In a multiple logistic
original article
Annals of Oncology
Table 4. Factors associated with the administration of inadequate analgesic therapy (PMI < 0) Covariates
P value
Multivariate OR (95% CI)
P value
1.00 0.96 (0.65–1.41)
0.8461
1.00 1.24 (0.82–1.89)
0.31
1.26 (0.47–3.55) 1.28 (0.47–3.49) 1.64 (0.60–4.51) 1.00 1.00 1.44 2.28 2.41 2.23 2.65 2.58
0.9477 0.9752 0.2135
1.72 (0.62–4.73) 1.78 (0.63–5.02) 1.89 (0.67–5.33) 1.00
Multivariate stepwise OR (95% CI)
P value
0.5813 0.4817 0.3428
(0.41–4.99) (0.89–5.84) (0.98–5.96) (0.85–5.90) (0.88–7.95) (1.00–6.63)
0.566 0.0872 0.0563 0.1043 0.082 0.0494
1.00 1.18 2.24 1.91 2.03 2.32 2.60
(0.33–4.22) (0.84–5.97) (0.75–4.89) (0.75–5.50) (0.76–7.08) (0.98–6.91)
0.7948 0.108 0.1755 0.1647 0.14 0.0554
1.00 1.66 (1.11–2.49)
0.0139
1.00 1.60 (1.04–2.44)
0.0317
1.00 1.52 (1.01–2.29)
0.0456
1.00 1.83 (1.23–2.70)
0.0026
1.00 1.75 (1.15–2.67)
0.0096
1.00 1.70 (1.14–2.55)
0.0096
PMI, Pain Management Index; OR, odds ratio; CI, confidence interval.
regression model, the same factors were independently associated with receiving an inadequate analgesic therapy (PMI < 0). Particularly, the admission to non-oncology units was associated with 75% increased risk (OR = 1.75, 95% CI = 1.15–2.67; P = 0.0096) and the absence of metastatic disease with 60% increased risk (OR = 1.60, 95% CI = 1.04–2.44; P = 0.0317), adjusted for other confounders. These findings were confirmed in a multivariate stepwise selection analysis (Table 4).
discussion In recent studies >50% of the hospitalized patients have been reported to present with moderate to severe pain in the previous 24 h, independent of the setting of inpatient care (oncology, internal medicine, surgery, orthopaedics, etc.) both in Italy [7–9] and in Europe [12], and under-treatment of cancer-related pain has been still recognized to be a significant finding in settings of care, supposed to be specialized in cancer pain management, such as anaesthesiology and/or oncology units from Italy [5]. However, to our knowledge, no data are yet available on the characteristics of analgesic therapy administered for cancer-related pain in patients admitted to different settings of care, namely oncology medical units and non-oncology units (surgery, internal medicine and orthopaedics wards). Our study provides the first comparative description of the differences in the attitude and modalities to use analgesic drugs for cancer patients treated in different settings of hospital care in Italy. Consistent with the results from other differently
Volume 21 | No. 10 | October 2010
designed studies, our study has first extended the notion that a significant number of patients may suffer from cancer-related pain despite receiving analgesic treatment, the proportions of patients with inadequately treated cancer-related pain being 11% and 19% in oncology and non-oncology units, respectively (Table 4). However, these percentages were definitely lower than those reported in the recent literature, being 25% and 43% in the studies by Apolone et al. [5] and Deandrea et al. [2], respectively. Such a difference could be related to a possible underestimation of the number of inadequately treated patients (PMI < 0), as our study has been designed to include and follow patients receiving analgesic therapy, so that those patients who may have suffered from pain, even with moderate to severe intensity, but not receiving analgesic therapy, may have been missed from our analysis. On the other hand, this difference could also be related to the fact that, while the other reported studies have tended to consider only the prescribed therapy, in our series the analgesic therapy actually administered to the patients, also including the rescue doses, has been considered and analysed. The clinical characteristics of our cancer patients admitted to the two settings of care to be compared are rather similar (Table 1) and, although patients with solid tumours account for the majority of the study population, haematological patients have also been encountered, consistent with the recently debated evidence, that, contrary to what has been reported in the past literature [13], a significant proportion of haematological patients (leukaemia and lymphoma) may suffer from pain not only in the last months of life [3] but also from the time of diagnosis [14]. As expected, patients with metastatic
doi:10.1093/annonc/mdq155 | 2091
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 19, 2015
Gender Male Female Age £64 65–74 75–84 ‡85 Primary cancer Breast Haematological Lung Gastrointestinal Genito-urinary Bone and soft tissues Others Metastasis Yes No Settings of care Oncology units Non-oncology units
Univariate OR (95% CI)
original article
2092 | Sichetti et al.
because they offer an appropriate standard of cancer pain treatment, which is superior to that found in non-oncology units. Institutional interventions designed to improve the assessment and treatment of pain in hospitalized cancer patients should be implemented and oncology units should be actively involved in these educational programmes.
acknowledgements ECAD Working Group consists of the following members—Alba (CN), Osp. S. Lazzaro: M. M. Avataneo, P. P. Fasolo, M. Rinaldi; Altamura (BA), Osp. Umberto I: M. C. Meleleo, R. Ricciardelli, R. Toputi; Ancona, INRCA: R. Ganzetti, F. Berre`, S. Spinicelli, M. Di Muzio; Atessa (CH), Osp. Vittorio Emanuele: P. D’Ovidio, A. Azzariti, Sr Margaret, Sr Clara, Sr Anna, Sr Isidora; Bari, AO Policlinico: M. Cristiano, M. E. Faggiano, M. Lattarulo; Bassano del Grappa (VI), PO di Bassano del Grappa: G. Tonato, M. S. Cogato, S. Zizzetti, G. Beltramello; Benevento, AO Rummo: A. Racca, A. Pasquariello, C. Goglia, V. Palmieri; Biella, Osp. degli Infermi: I. Marone, S. Terziotti, M. R. Fogliano, L. Lanzone, A. Penna, N. Anino, R. Butta`; Bra (CN), Osp. S. Spirito: L. Infante, L. Castellino, A. Vernassa, B. Racca, L. Bergesio, G. Serra; Bussolengo (VR), PO di Bussolengo: V. Bertasi, L. Masconale, A. Ballarin, M. Cevese, L. Furioni; Camaiore (LU), Osp. Versilia: D. Musetti, S. Nencioni, F. Dolfi, F. Giuntoli, B. Castori; Camposampiero (PD), PO di Camposampiero: A. Pedrini, L. Cordiano, C. Paolello, M. Cavinato; Carpi (MO), Osp. Ramazzini: A. Delbue, P. Baffoni, S. Sanguanini, R. Prandi; Casoli (CH), PO Consalvi: F. De Vita, L. Rinaldi, S. Rossi, A. Di Medio, P. Di Cino, G. De Angelis; Chieri Moncalieri Carmagnola (TO), Osp. ASL TO5: E. Abbate, G. Tartaglia, N. Morello, C. Grillo, S. Vendemmiati, A. Dainese; Chieti, PO Clinicizzato SS. Annunziata: N. Di Cicco, F. Margiotta, E. Angelucci, M. Cieri; Como, Osp. S. Anna: P. Ardovino, M. Carughi, E. Pizzuti, M. Giordano, C. Longhi, M. C. Valli, P. Pugliese, N. Arnaboldi, A. Scanziani, D. Livio; Erice Casasanta (TP), AO S. Antonio Abate: R. Sanci, G. Giacalone, F. Basciano, G. Castello, V. Spano`, A. Donato, G. Agosta; Fabriano (AN), Osp. Civile E. Profili: A. M. Ruggieri, C. De Vito, M. Candela, L. Salari; Figline Valdarno (FI), Osp. Serristori: L. Bencivenni, P. Fratini, L. Bruschetini; Firenze, Casa di cura Villa Donatello: G. Paggi, A. Vannucci, S. Galli, G. Bagnoli; Firenze, Casa di cura Villanova: G. Paggi, V. De Leonardis, M. Balloni; Firenze, IOT Palagi: L. Pazzagli, G. Croppi, L. Panarese, S. Falchi, A. Cianciullo, G. Cioni; Galatina (LE), Osp. S. Caterina Novella: R. Vergine, P. De Pascalis, G. Gemma, S. Maghenzani; Gallarate (VA), AO S. Antonio Abate: G. Monina, M. Ghiringhelli, E. Albini, R. Riva, P. F. Interdonato, R. Fonzo; Ialmicco (UD), Osp. di Ialmicco: G. Millevoi, F. La Ferla, R. Maricchio; Ivrea (TO), Osp. Civile: L. Rocatti, P. Abrate, S. Bretti, G. Regis; Lanciano (CH), Osp. Renzetti: C. Di Fabio, G. Petragnani, R. Di Tommaso; Milano, IEO: P. Paochi, E. Omodeo Sale`, V. Sirna, G. Magon; Mirano (VE), Ospedali Azienda ULSS 13 Veneto: S. Zardo, A. Valenti, S. Cognolato, G. Azzarello, B. Silvestri, R. Dona`, M. Coro`, S. Masiero; Modena, Osp. USL di Modena: L. Buonaccorso, R. Guerzoni, D. Bolognesi, E. Bandieri, F. Artioli; Mondovı` (CN), Presidio
Volume 21 | No. 10 | October 2010
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 19, 2015
disease have more frequently been admitted to oncology units, where also younger patients with potentially treatable tumours, such as breast and haematologic cancers, have been significantly more represented, compared with non-oncology units (Table 1). Clearly, the most striking finding from our study is the differential use of analgesic drugs in the two settings of care, the use of non-opioids being more frequent in non-oncology units, while, conversely, the use of strong opioids being more frequent in oncology units (Figure 1). When the administration of different analgesic drugs has been analysed in relation to the intensity of pain and the adequacy of pain management evaluated by the PMI assessment, again, the different behaviours between the two settings of care have resulted to be impressively different, the use of strong opioids for the treatment of severe pain being significantly less frequent in non-oncology units, where, conversely the use of non-opioids for the treatment of severe pain has resulted significantly more frequently than that in oncology units (Table 3). Few studies have shown that among the different factors potentially associated with the inadequacy of analgesic therapy, the typology of the setting of care may be highly relevant, although, only different clinical settings have so far been compared, such as hospitals with hospices, by evaluating clinically heterogeneous patient groups [2, 5]. Our study, by comparing two rather homogeneous cancer patient populations treated within the same hospital, has provided the first demonstration that admission to a non-oncology unit and the absence of metastasis are factors significantly and independently associated with an increased risk to receive an inadequate treatment, as also confirmed by a multivariate analysis (Table 4). The results of this survey do not allow to investigate the multiple causes of under-treatment of pain, but certainly opiophobia is a key factor, especially in the non-oncology settings. Opiophobia describes the fear of using opioid drugs associated with a lack of knowledge about appropriate analgesic drug treatment, complicated by an overwhelming fear of abuse and addiction [15]. It results in an under-treatment because the opioid is not prescribed at all or prescribed with inadequate doses or routes of administration. Opiophobia may be less pronounced when taking care of patients with metastatic disease, favouring an increased and possibly more adequate use of strong opioids in this poor-prognosis patients group. Opiophobia may also be due to the reluctance of the patients and/or their family in accepting strong opioids [16–18] and result in a low compliance to the recommended doses. Moreover, the prescription of the various opioids may also be more influenced by marketing than by evidences produced by comparative studies [19]. On the other hand, in the oncology setting of care, much efforts have been made in the last few years to implement educational programmes in the management of cancer-related pain and palliative medicine, which have allowed to train oncologists with a palliative care competence and attitude, working within each single oncology unit or devoted to a consultant activities. In conclusion, the results of this study have shown that, actually, oncology units are the best settings of care for patients with cancer-related pain, especially with metastatic disease,
Annals of Oncology
Annals of Oncology
references 1. Van den Beuken-van Everdingen MHJ, De Rijke JM, Kessels AG et al. Prevalence of pain in patients with cancer: a systematic review of the past 40 years. Ann Oncol 2007; 18: 1437–1449. 2. Deandrea S, Montanari M, Moja L et al. Prevalence of undertreatment in cancer pain. A review of published literature. Ann Oncol 2008; 19(12): 1985–1991.
Volume 21 | No. 10 | October 2010
3. Costantini M, Ripamonti C, Beccaro M et al. Prevalence, distress, management and relief of pain during the last three months of cancer patients’ life. Results of an Italian mortality follow-back survey. Ann Oncol 2009; 20: 729–735. 4. Breivik H, Cherny N, Collett F et al. Cancer-related pain: a pan-European survey of prevalence, treatment, and patient attitudes. Ann Oncol 2009; 20: 1420–1433. 5. Apolone G, Corli O, Caraceni A et al. Pattern and quality of care of cancer pain management. Results from the Cancer Pain Outcome Research Study Group. Br J Cancer 2009; 100: 1566–1574. 6. Cleeland CS, Gonin R, Hatfield AK et al. Pain and its treatment in outpatients with metastatic cancer. N Engl J Med 1994; 330: 592–596. 7. Costantini M, Viterboni P, Flego G. Prevalence of pain in Italian hospitals: results of a regional cross-sectional survey. J Pain Symptom Manage 2002; 23: 221–230. 8. Ripamonti C, Zecca E, Brunelli C et al. Pain experienced by patients hospitalized at the national Cancer Institute of Milan: research project ‘Towards a pain-free hospital’. Tumori 2000; 86: 412–418. 9. Visentin M, Zanolin E, Trentin L et al. Prevalence and treatment of pain in adults admitted to Italian hospital. Eur J Pain 2005; 9: 61–67. 10. Sichetti D, Di Biagio K, Anino N et al. Il Modello di sorveglianza ECADO—Epidemiologia Clinico-Assistenziale del Dolore in Osdpedale: il contributo degli infermieri. Assistenza infermieristica e ricerca 2009; 28(2): 73–81. 11. World Health Organization. Cancer Pain Relief. 2nd edition. Geneva, Switzerland: World Health Organization, 1996. 12. Holtan A, Aass N, Nordøy T et al. Prevalence of pain in hospitalised cancer patients in Norway: a national survey. Palliat Med 2007; 21: 7–13. 13. Foley KM, Kathleen M. The treatment of cancer pain. New Engl J Med 1985; 313: 84–95. 14. Morselli M, Bandieri E, Zanin R et al. Pain and emotional distress in leukemia patients at diagnosis. Leuk Res 2010; 34: e67–e68. 15. Zenz M, Willweber-Strumpf A. Opiophobia and cancer pain in Europe. Lancet 1993; 341: 1075–1076. 16. Paice JA, Toy C, Shott S. Barriers to cancer pain relief: fear of tolerance and addiction. J Pain Symptom Manage 1998; 16: 1–9. 17. Weiss SC, Emanuel LL, Fairclough DL, Emanuel EJ. Understanding the experience of pain in terminally ill patients. Lancet 2001; 357: 1311–1315. 18. Ward SE, Goldberg N, Miller-McCauley M et al. Patient-related barriers to management of cancer pain. Pain 1993; 52: 319–324. 19. Bandieri E, Chirarolanza A, Luppi M et al. Prescription of opioids in Italy: everything but the morphine [letters to the editor]. Ann Oncol 2009; 20: 961–962.
doi:10.1093/annonc/mdq155 | 2093
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 19, 2015
ASL 16 Piemonte: A. Bramardi, B. Bovetti, L. Bagnasco; Monselice (PD), PO ULSS 17 di Este: R. Congedo, A. Amato, M. Cogo, F. Visentini, G. Reffo, F. Bellamio, N. Marconcini; Montebelluna (TV), Ospedali Azienda ULSS 3 Veneto: R. Callegari, M. Osti, M. Ragazzi, G. Cecchetto, L. Tessaro, L. Lusiani, S. Barichello, F. Sartor, M. Cervi, G. Rech, F. Patassini, M. Gottardo; Montecchio Emilia (RE), Osp. E. Franchini: S. Fietta, G. Pagliarini, A. Cocconi, R. Aldini; Palermo, AO Civico: P. Finocchiaro, P. Marrone, G. Trizzino, M. Di Liberto, M. Tamburo De Bella; Pescara, Osp. Civile S. Spirito: F. Sanita`, M. Di Staso, M. Lombardo, G. Gidaro, P. Lelli Chiesa, G. Parruti, L. Ceccomancini, I. Pelusi; Pisa, AOU Pisana: M. Polvani, F. Vivaldi, A. Paolicchi, M. Marcaccini, C. Venturi; Ragusa, AO Civile Maria Paterno` Arezzo: M. Martorana; Roma, AOU Policlinico Tor Vergata: M. G. Celeste, A. Marra, L. Pacca, G. Visconti, A. Gatti, C. Cappitella; Rovereto (TN), Osp. S. Maria del Carmine: G. Dusi, M. Mattarei, R. Girardello, R. Falzone, N. Stefenelli, M. Saiani, S. Scarabello Vettore; Senigallia (AN), Osp. di Senigallia: P. Borona, R. Connestari, V. M. Carfi; Taranto, AO SS. Annunziata: R. Moscogiuri, C. A. Pennetta, D. Zollino, D. Semeraro, G. Caldaralo, M. A. De Maria; Teramo, Osp. Mazzini: M. Turchetti, I. Senesi, O. Cipolletti, S. Erpini, P. Lisciani; Tradate (VA), Osp. Galmarini: P. Lusuriello, A. Volonte`, M. D. Centemeri; Trento, Osp. di Trento: M. Boni, A. Campomori, F. Branz, E. Geat, M. Brugnaria, E. Modena; Vasto (CH), Osp. S. Pio Da Pietralcina: L. Di Fabio, C. Spoltore, E. Testa; Vercelli, Osp. Sant’Andrea: M. Giolito, M. Cianfanelli.
original article
Annals of Oncology 21: 2088–2093, 2010 doi:10.1093/annonc/mdq155 Published online 31 March 2010
Impact of setting of care on pain management in patients with cancer: a multicentre cross-sectional study D. Sichetti1 , E. Bandieri2 , M. Romero1 , K. Di Biagio1, M. Luppi3*, M. Belfiglio1, G. Tognoni1 & C. I. Ripamonti4* for ECAD Working Group 1
Department of Clinical Pharmacology and Epidemiology, Consorzio Mario Negri Sud, Santa Maria Imbaro (Chieti); 2Palliative Care Unit, Ausl Modena, Modena; Department of Oncology, Haematology and Respiratory Diseases, Azienda Ospedaliera Universitaria, University of Modena and Reggio Emilia, Modena; 4Supportive Care in Cancer Unit, IRCCS Foundation, National Cancer Institute, Milano, Italy 3
Background: No study has so far addressed whether differences do exist in the management of cancer-related pain in patients admitted to oncology and non-oncology settings. Patients and methods: A multicentre cross-sectional study in 48 Italian hospitals has enrolled 819 patients receiving analgesic therapy for cancer-related pain. Demographics and clinical and analgesic therapy information have been prospectively collected by standardized forms. Adequacy of pain management has been evaluated by the Pain Management Index (PMI). Results: Differences in the analgesic drug administration according to settings of care have been evident, non-opioids more frequently being administered in non-oncology units (19.6% versus 7.0%; P < 0.0001), while strong opioids are more frequently used in the oncology units (69.5% versus 51.9%; P < 0.0001). The number of patients receiving inadequate therapy (PMI < 0) has lowered in oncology compared with non-oncology units (11.3% versus 18.8%; P = 0.0024). Results of multiple logistic regression analysis have shown that the admission to non-oncology setting [odds ratio (OR) = 1.75, 95% confidence interval (CI) = 1.15–2.67; P = 0.0096] and the absence of metastatic disease (OR = 1.60, 95% CI = 1.04–2.44; P = 0.0317) were independent factors associated with an increased risk of receiving an inadequate analgesic therapy. Conclusion: Oncology wards provide the most adequate standard of analgesic therapy for cancer-related pain. Key words: analgesic drugs, neoplasms, pain, settings of care
introduction Pain is one of the most feared and burdensome symptoms in cancer patients and unrelieved pain continues to be a substantial worldwide public health concern. Consistent with this, a systematic review of the literature investigated the prevalence of pain in different disease stages and types of cancer during the period 1966–2005 [1]. The pooled data from 52 articles showed that pain prevalence was 64% in patients with metastatic, advanced or terminal phase disease, 59% in patients on anticancer treatment and 33% in patients after curative treatment. No difference in pain prevalence was found *Correspondence to: Prof. M. Luppi, Department of Oncology, Hematology and Respiratory Diseases, Azienda Ospedaliera Universitaria, University of Modena and Reggio Emilia, Modena, Italy. Tel: +39-059-4225570; Fax: +39-059-4224549; E-mail: [email protected] or Dr. C. I. Ripamonti, Supportive Care in Cancer Unit, IRCCS Foundation, National Cancer Institute, Via Venezian n 1, 20133 Milano, Italy. Tel: +39-02-2390-3640; Fax: +39-02-2390-3656; E-mail: [email protected]
These authors contributed equally.
between the patients during anticancer treatment and those in advanced or terminal phase of the disease. In the 18 studies reporting the pain severity, one-third of the patients rated their pain as moderate to severe [1]. Moreover, in the systematic review of the literature carried out by Deandrea et al. [2] on studies published from 1994 to 2007, nearly half of the cancer patients were under-treated with a high variability across study designs and clinical settings. Recent studies conducted both in Italy and in Europe [3, 4] confirmed these data, showing that pain was present in all phases of cancer disease (early and metastatic) and was not adequately treated in a significant percentage of patients, ranging from 56% to 82.3%. In particular, Apolone et al. [5] evaluated prospectively the adequacy of analgesic care of cancer patients by means of Pain Management Index (PMI) [6] in 1802 valid cases of in- and outpatients with advanced/metastatic solid tumour enrolled in 110 centres specifically devoted to cancer and/or pain management (oncology/pain/palliative centres or hospices). The study showed that patients were still classified as potentially under-treated in 25.3% of the cases (range 9.8%–55.3%).
ª The Author 2010. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected]
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 19, 2015
original article
Received 1 March 2010; accepted 4 March 2010
original article
Annals of Oncology
patients and methods study design and setting A multicentre cross-sectional study has been carried out in 48 hospitals, located in 13 different Italian regions, to identify patients receiving analgesics. In each hospital, a multidisciplinary working group, made up of physicians, nurses and pharmacists, has been established. Each local working group, promoting this survey as a hospital pharmaco-surveillance activity, has predetermined index days to carry out the survey and coordinated participant wards. Overall, in the 164 clinical wards involved, all patients treated with analgesic therapy, irrespective of age and types of pain, have been considered eligible for this project. The patients, treated with an analgesic drug according to the usual practice, have been sampled in six index days (from 12 February 2007 through 29 July 2008). Wards have been divided into two area-type clusters, namely oncology and non-oncology wards, the latter including surgery, internal medicine and orthopaedics wards. This project has been authorized by the Hospital General Management and by the Local Ethical Committee. For each patient included in the study, data have been obtained according to Italian law about privacy (D.Lgs. 196/2003), so that all patients have been recruited anonymously to the central data analysis.
data collection Information regarding patients and wards have been collected using standardized forms. For each patient involved in the survey, demographics (age and gender), clinical treatment (type of pain, primary cancer and presence of metastasis) and analgesic treatment (type of analgesic drug prescribed and administered in the last 24 h, route of administration, dosage and adjuvant treatment modality) have been obtained from medical records. Patients have been interviewed by a pharmacist—considered an independent health professional not involved in patients’ care—concerning their perception of pain control and their pain intensity at interview and the worst intensity of pain perceived during the previous 24 h. Pain intensity has been estimated by using a four-level Verbal Rating Scale as no pain,
Volume 21 | No. 10 | October 2010
mild pain, moderate pain and severe pain. We have defined and scored pain levels as follows—0 points: no pain; 1 point: mild pain; 2 points: moderate pain and 3 points: severe pain. Health care professionals (nurses and/or clinicians) have been asked about their perception on the effectiveness of the analgesic treatment administered and the possible reasons for the absence of treatment response. Moreover, patients have also been asked about their relationship with health care professionals, and, specifically, whether they have communicated their pain, whether they have received a treatment after communication and the longest time interval they have had to wait before the initiation of treatment. Finally, patients have also evaluated the efficacy of the analgesic therapy received. Reasons for failed interviews (refusal, patient’s absence, cognitive limitations/impairment, language barrier or other reasons) have been collected by the interviewer.
pain management index To evaluate the adequacy of pain management according to the World Health Organization (WHO) guidelines [11], PMI has been assessed [6]. The Cleeland’s PMI compares the analgesic drug with the level of pain reported by the patients and is computed by subtracting pain scores from analgesic scores [6]. The analgesics were scored as follows: 0 points = no analgesic; 1 point = WHO I step (non-opioid); 2 points = WHO II step (weak opioids, such as codeine and tramadol) and 3 points = WHO III step (strong opioids, such as morphine, oxycodone, methadone, fentanyl, hydromorphone and buprenorphine). We have defined and scored pain levels as follows: 0 points = no pain; 1 point = mild pain; 2 points = moderate pain and 3 points = severe pain. PMI ranges from 23 (patients with severe pain receiving no drug at all) to +3 (patients receiving strong opioids and reporting pain relief). Negative scores indicate inadequate orders for analgesic drugs and have been considered as under-treatment, whereas scores of 0 points and higher have been considered as indicators of acceptable treatment.
statistical analysis Patients’ baseline characteristics (gender, age, primary cancer and presence of metastasis), analgesic treatment (non-opioids, weak opioids and strong opioids) and (in)adequacy of cancer pain management (computed using PMI) have been reported as percentages in the two different settings of care (oncology and non-oncology wards) and have been compared with chi-square test. For univariate analysis, the following variables were considered: gender, age, primary cancer, metastasis and setting of care. In order to identify independent factors associated with inadequate pain treatment (PMI < 0), a multiple logistic regression model has been used to control for the same variables tested in the univariate analysis, simultaneously. Multivariate models were also constructed using stepwise selection methods. The results have been expressed as odds ratios (ORs) and 95% confidence intervals (95% CIs). P values <0.05 have been considered significant. All analyses have been carried out using SAS Statistical Package Release 9.2 (SAS Institute, Cary, NC).
results A total of 819 patients were included in the analysis. Of them, 55.4% were older than 65 years and 51.6% of them were male. The characteristics of the patients, 426 in oncology and 393 in non-oncology units, grouped by setting of care are reported in Table 1. The two patient groups considered have shown statistically significant differences (P < 0.001) in relation to
doi:10.1093/annonc/mdq155 | 2089
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 19, 2015
Moreover, different studies [7–9] have clearly shown a lack of adequate pain management in the majority of the Italian hospitals and have reinforced the need for the implementation of organizational and educational interventions, in order to emend this widespread malpractice. However, no study has so far been carried out in Italy with the aim to compare the modalities of treatment of cancerrelated pain in patients admitted to different speciality settings of care. The ECAD-O (Epidemiologia Clinico-Assistenziale del Dolore in Ospedale) (Clinical-Caring Epidemiology of Pain in Hospital) group has been established to activate a surveillance on pain management, as part of a routine care, by creating a multicentre network of hospitals and a multidisciplinary working group. Specifically, a survey has been carried out in order to investigate the characteristics of inpatients treated with analgesic drugs and to evaluate the perception of the level of pain control both from patients and health care professionals [10]. The preliminary results of this ECAD-O survey [10] have shown that cancer pain remains one of the main causes of analgesic therapy administration both in oncology and in non-oncology (surgery, internal medicine and orthopaedics) wards. The aim of the present research study was to determine the impact of setting of care on pain management in patients with cancer.
original article
Annals of Oncology
certain clinical features: age, primary cancer and metastatic disease (Table 1). The types of analgesic therapy, including non-opioid, weak opioid and strong opioid drugs administered to the patients in the two different settings of care, have also been analysed and differences between the two settings have are evident. In particular, as reported in Figure 1, the difference was statistically significant for non-opioid drugs that have been administered more frequently in non-oncology than in oncology units (19.6% versus 7.0%; P < 0.0001) and for the use of strong opioids that has resulted more frequent in the oncology setting (69.5% versus 51.9%; P < 0.0001). The frequencies of different types of analgesic therapy, administered to the two patient groups, following the evaluation of the intensity of pain recorded in the last 24 h
Table 1. Characteristics of the patients in the two settings of care
Oncology units (n = 426), n (%)
P value
180 (45.8) 213 (54.2)
216 (50.7) 210 (49.3)
151 101 107 34
214 141 62 9
0.1607
<0.0001 (38.4) (25.7) (27.2) (8.7)
(50.2) (33.1) (14.6) (2.1) <0.0001
19 (4.9) 72 (18.4) 135 (34.5) 60 (15.4) 30 (7.7) 30 (7.7) 45 (11.5) 2
30 (7.1) 79 (18.6) 90 (21.2) 60 (14.2) 22 (5.2) 52 (12.3) 91 (21.5) 2
146 (37.9) 239 (62.1) 8
98 (23.1) 326 (76.9) 2
<0.0001
Table 2. Distribution of the patients according to the analgesic treatment received and pain intensity in the two settings of care Pain 24 h
Non-oncology units None Mild Moderate Severe Total Oncology units None Mild Moderate Severe Total
Analgesic treatment received Strong Weak Nonopioids, opioids, opioids, n n n
Total
12 18 21a 26a 77
28 25 32 27a 112
22 46 67 69 204
62 89 120 122 393
7 8 9a 6a 30
14 25 28 33a 100
51 72 68 105 296
72 105 105 144 426
a
Patients with inadequate treatment according to PMI. PMI, Pain Management Index.
Table 3. Distribution of patients according to the adequacy/inadequacy of the analgesic treatment received and the pain intensity as reported by the patients (PMI) in the two settings of care
PMI <0 (inadequate) ‡0 (adequate) Figure 1. Analgesic treatment received in the two settings of care.
2090 | Sichetti et al.
Non-oncology units (n =393), n (%)
Oncology units (n =426), n (%)
74 (18.8) 319 (81.2)
48 (11.3) 378 (88.7)
P value
0.0024
PMI, Pain Management Index.
Volume 21 | No. 10 | October 2010
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 19, 2015
Gender Female Male Age £64 65—74 75—84 ‡85 Primary cancer Haematological Lung Gastrointestinal Genito-urinary Bone and soft tissues Breast Others Missing Metastasis No Yes Missing
Setting of care Non-oncology units (n = 393), n (%)
before the interview, have also been compared. As reported in Table 2, strong opioids have been used in 69 of 122 (56.5%) patients with severe pain in non-oncology units and in 105 of 144 (72.9%) patients with pain of similar intensity in oncology units. Non-opioids have been used in 26 patients (21.3%) with severe pain in non-oncology units and in only 6 (4.2%) patients in oncology units. Moreover, strong opioids have also been used for the treatment of mild pain in 46 (51.7%) and 72 (68.6%) patients in non-oncology and oncology units, respectively. By applying the PMI criteria to our study population, the number of patients receiving an appropriate treatment (PMI ‡ 0) has increased in oncology setting (88.7%) than in nononcology setting (81.2%). On the contrary, inadequate therapy (PMI < 0) has been administered to a lower number of patients in oncology units (11.3%) compared with non-oncology units (18.8%). These differences were statistically significant (P = 0.0024; Table 3). Univariate analysis showed that the admission to nononcology units and the absence of metastatic disease showed a significant relationship with respect to receiving an inadequate analgesic treatment (Table 4). In a multiple logistic
original article
Annals of Oncology
Table 4. Factors associated with the administration of inadequate analgesic therapy (PMI < 0) Covariates
P value
Multivariate OR (95% CI)
P value
1.00 0.96 (0.65–1.41)
0.8461
1.00 1.24 (0.82–1.89)
0.31
1.26 (0.47–3.55) 1.28 (0.47–3.49) 1.64 (0.60–4.51) 1.00 1.00 1.44 2.28 2.41 2.23 2.65 2.58
0.9477 0.9752 0.2135
1.72 (0.62–4.73) 1.78 (0.63–5.02) 1.89 (0.67–5.33) 1.00
Multivariate stepwise OR (95% CI)
P value
0.5813 0.4817 0.3428
(0.41–4.99) (0.89–5.84) (0.98–5.96) (0.85–5.90) (0.88–7.95) (1.00–6.63)
0.566 0.0872 0.0563 0.1043 0.082 0.0494
1.00 1.18 2.24 1.91 2.03 2.32 2.60
(0.33–4.22) (0.84–5.97) (0.75–4.89) (0.75–5.50) (0.76–7.08) (0.98–6.91)
0.7948 0.108 0.1755 0.1647 0.14 0.0554
1.00 1.66 (1.11–2.49)
0.0139
1.00 1.60 (1.04–2.44)
0.0317
1.00 1.52 (1.01–2.29)
0.0456
1.00 1.83 (1.23–2.70)
0.0026
1.00 1.75 (1.15–2.67)
0.0096
1.00 1.70 (1.14–2.55)
0.0096
PMI, Pain Management Index; OR, odds ratio; CI, confidence interval.
regression model, the same factors were independently associated with receiving an inadequate analgesic therapy (PMI < 0). Particularly, the admission to non-oncology units was associated with 75% increased risk (OR = 1.75, 95% CI = 1.15–2.67; P = 0.0096) and the absence of metastatic disease with 60% increased risk (OR = 1.60, 95% CI = 1.04–2.44; P = 0.0317), adjusted for other confounders. These findings were confirmed in a multivariate stepwise selection analysis (Table 4).
discussion In recent studies >50% of the hospitalized patients have been reported to present with moderate to severe pain in the previous 24 h, independent of the setting of inpatient care (oncology, internal medicine, surgery, orthopaedics, etc.) both in Italy [7–9] and in Europe [12], and under-treatment of cancer-related pain has been still recognized to be a significant finding in settings of care, supposed to be specialized in cancer pain management, such as anaesthesiology and/or oncology units from Italy [5]. However, to our knowledge, no data are yet available on the characteristics of analgesic therapy administered for cancer-related pain in patients admitted to different settings of care, namely oncology medical units and non-oncology units (surgery, internal medicine and orthopaedics wards). Our study provides the first comparative description of the differences in the attitude and modalities to use analgesic drugs for cancer patients treated in different settings of hospital care in Italy. Consistent with the results from other differently
Volume 21 | No. 10 | October 2010
designed studies, our study has first extended the notion that a significant number of patients may suffer from cancer-related pain despite receiving analgesic treatment, the proportions of patients with inadequately treated cancer-related pain being 11% and 19% in oncology and non-oncology units, respectively (Table 4). However, these percentages were definitely lower than those reported in the recent literature, being 25% and 43% in the studies by Apolone et al. [5] and Deandrea et al. [2], respectively. Such a difference could be related to a possible underestimation of the number of inadequately treated patients (PMI < 0), as our study has been designed to include and follow patients receiving analgesic therapy, so that those patients who may have suffered from pain, even with moderate to severe intensity, but not receiving analgesic therapy, may have been missed from our analysis. On the other hand, this difference could also be related to the fact that, while the other reported studies have tended to consider only the prescribed therapy, in our series the analgesic therapy actually administered to the patients, also including the rescue doses, has been considered and analysed. The clinical characteristics of our cancer patients admitted to the two settings of care to be compared are rather similar (Table 1) and, although patients with solid tumours account for the majority of the study population, haematological patients have also been encountered, consistent with the recently debated evidence, that, contrary to what has been reported in the past literature [13], a significant proportion of haematological patients (leukaemia and lymphoma) may suffer from pain not only in the last months of life [3] but also from the time of diagnosis [14]. As expected, patients with metastatic
doi:10.1093/annonc/mdq155 | 2091
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 19, 2015
Gender Male Female Age £64 65–74 75–84 ‡85 Primary cancer Breast Haematological Lung Gastrointestinal Genito-urinary Bone and soft tissues Others Metastasis Yes No Settings of care Oncology units Non-oncology units
Univariate OR (95% CI)
original article
2092 | Sichetti et al.
because they offer an appropriate standard of cancer pain treatment, which is superior to that found in non-oncology units. Institutional interventions designed to improve the assessment and treatment of pain in hospitalized cancer patients should be implemented and oncology units should be actively involved in these educational programmes.
acknowledgements ECAD Working Group consists of the following members—Alba (CN), Osp. S. Lazzaro: M. M. Avataneo, P. P. Fasolo, M. Rinaldi; Altamura (BA), Osp. Umberto I: M. C. Meleleo, R. Ricciardelli, R. Toputi; Ancona, INRCA: R. Ganzetti, F. Berre`, S. Spinicelli, M. Di Muzio; Atessa (CH), Osp. Vittorio Emanuele: P. D’Ovidio, A. Azzariti, Sr Margaret, Sr Clara, Sr Anna, Sr Isidora; Bari, AO Policlinico: M. Cristiano, M. E. Faggiano, M. Lattarulo; Bassano del Grappa (VI), PO di Bassano del Grappa: G. Tonato, M. S. Cogato, S. Zizzetti, G. Beltramello; Benevento, AO Rummo: A. Racca, A. Pasquariello, C. Goglia, V. Palmieri; Biella, Osp. degli Infermi: I. Marone, S. Terziotti, M. R. Fogliano, L. Lanzone, A. Penna, N. Anino, R. Butta`; Bra (CN), Osp. S. Spirito: L. Infante, L. Castellino, A. Vernassa, B. Racca, L. Bergesio, G. Serra; Bussolengo (VR), PO di Bussolengo: V. Bertasi, L. Masconale, A. Ballarin, M. Cevese, L. Furioni; Camaiore (LU), Osp. Versilia: D. Musetti, S. Nencioni, F. Dolfi, F. Giuntoli, B. Castori; Camposampiero (PD), PO di Camposampiero: A. Pedrini, L. Cordiano, C. Paolello, M. Cavinato; Carpi (MO), Osp. Ramazzini: A. Delbue, P. Baffoni, S. Sanguanini, R. Prandi; Casoli (CH), PO Consalvi: F. De Vita, L. Rinaldi, S. Rossi, A. Di Medio, P. Di Cino, G. De Angelis; Chieri Moncalieri Carmagnola (TO), Osp. ASL TO5: E. Abbate, G. Tartaglia, N. Morello, C. Grillo, S. Vendemmiati, A. Dainese; Chieti, PO Clinicizzato SS. Annunziata: N. Di Cicco, F. Margiotta, E. Angelucci, M. Cieri; Como, Osp. S. Anna: P. Ardovino, M. Carughi, E. Pizzuti, M. Giordano, C. Longhi, M. C. Valli, P. Pugliese, N. Arnaboldi, A. Scanziani, D. Livio; Erice Casasanta (TP), AO S. Antonio Abate: R. Sanci, G. Giacalone, F. Basciano, G. Castello, V. Spano`, A. Donato, G. Agosta; Fabriano (AN), Osp. Civile E. Profili: A. M. Ruggieri, C. De Vito, M. Candela, L. Salari; Figline Valdarno (FI), Osp. Serristori: L. Bencivenni, P. Fratini, L. Bruschetini; Firenze, Casa di cura Villa Donatello: G. Paggi, A. Vannucci, S. Galli, G. Bagnoli; Firenze, Casa di cura Villanova: G. Paggi, V. De Leonardis, M. Balloni; Firenze, IOT Palagi: L. Pazzagli, G. Croppi, L. Panarese, S. Falchi, A. Cianciullo, G. Cioni; Galatina (LE), Osp. S. Caterina Novella: R. Vergine, P. De Pascalis, G. Gemma, S. Maghenzani; Gallarate (VA), AO S. Antonio Abate: G. Monina, M. Ghiringhelli, E. Albini, R. Riva, P. F. Interdonato, R. Fonzo; Ialmicco (UD), Osp. di Ialmicco: G. Millevoi, F. La Ferla, R. Maricchio; Ivrea (TO), Osp. Civile: L. Rocatti, P. Abrate, S. Bretti, G. Regis; Lanciano (CH), Osp. Renzetti: C. Di Fabio, G. Petragnani, R. Di Tommaso; Milano, IEO: P. Paochi, E. Omodeo Sale`, V. Sirna, G. Magon; Mirano (VE), Ospedali Azienda ULSS 13 Veneto: S. Zardo, A. Valenti, S. Cognolato, G. Azzarello, B. Silvestri, R. Dona`, M. Coro`, S. Masiero; Modena, Osp. USL di Modena: L. Buonaccorso, R. Guerzoni, D. Bolognesi, E. Bandieri, F. Artioli; Mondovı` (CN), Presidio
Volume 21 | No. 10 | October 2010
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 19, 2015
disease have more frequently been admitted to oncology units, where also younger patients with potentially treatable tumours, such as breast and haematologic cancers, have been significantly more represented, compared with non-oncology units (Table 1). Clearly, the most striking finding from our study is the differential use of analgesic drugs in the two settings of care, the use of non-opioids being more frequent in non-oncology units, while, conversely, the use of strong opioids being more frequent in oncology units (Figure 1). When the administration of different analgesic drugs has been analysed in relation to the intensity of pain and the adequacy of pain management evaluated by the PMI assessment, again, the different behaviours between the two settings of care have resulted to be impressively different, the use of strong opioids for the treatment of severe pain being significantly less frequent in non-oncology units, where, conversely the use of non-opioids for the treatment of severe pain has resulted significantly more frequently than that in oncology units (Table 3). Few studies have shown that among the different factors potentially associated with the inadequacy of analgesic therapy, the typology of the setting of care may be highly relevant, although, only different clinical settings have so far been compared, such as hospitals with hospices, by evaluating clinically heterogeneous patient groups [2, 5]. Our study, by comparing two rather homogeneous cancer patient populations treated within the same hospital, has provided the first demonstration that admission to a non-oncology unit and the absence of metastasis are factors significantly and independently associated with an increased risk to receive an inadequate treatment, as also confirmed by a multivariate analysis (Table 4). The results of this survey do not allow to investigate the multiple causes of under-treatment of pain, but certainly opiophobia is a key factor, especially in the non-oncology settings. Opiophobia describes the fear of using opioid drugs associated with a lack of knowledge about appropriate analgesic drug treatment, complicated by an overwhelming fear of abuse and addiction [15]. It results in an under-treatment because the opioid is not prescribed at all or prescribed with inadequate doses or routes of administration. Opiophobia may be less pronounced when taking care of patients with metastatic disease, favouring an increased and possibly more adequate use of strong opioids in this poor-prognosis patients group. Opiophobia may also be due to the reluctance of the patients and/or their family in accepting strong opioids [16–18] and result in a low compliance to the recommended doses. Moreover, the prescription of the various opioids may also be more influenced by marketing than by evidences produced by comparative studies [19]. On the other hand, in the oncology setting of care, much efforts have been made in the last few years to implement educational programmes in the management of cancer-related pain and palliative medicine, which have allowed to train oncologists with a palliative care competence and attitude, working within each single oncology unit or devoted to a consultant activities. In conclusion, the results of this study have shown that, actually, oncology units are the best settings of care for patients with cancer-related pain, especially with metastatic disease,
Annals of Oncology
Annals of Oncology
references 1. Van den Beuken-van Everdingen MHJ, De Rijke JM, Kessels AG et al. Prevalence of pain in patients with cancer: a systematic review of the past 40 years. Ann Oncol 2007; 18: 1437–1449. 2. Deandrea S, Montanari M, Moja L et al. Prevalence of undertreatment in cancer pain. A review of published literature. Ann Oncol 2008; 19(12): 1985–1991.
Volume 21 | No. 10 | October 2010
3. Costantini M, Ripamonti C, Beccaro M et al. Prevalence, distress, management and relief of pain during the last three months of cancer patients’ life. Results of an Italian mortality follow-back survey. Ann Oncol 2009; 20: 729–735. 4. Breivik H, Cherny N, Collett F et al. Cancer-related pain: a pan-European survey of prevalence, treatment, and patient attitudes. Ann Oncol 2009; 20: 1420–1433. 5. Apolone G, Corli O, Caraceni A et al. Pattern and quality of care of cancer pain management. Results from the Cancer Pain Outcome Research Study Group. Br J Cancer 2009; 100: 1566–1574. 6. Cleeland CS, Gonin R, Hatfield AK et al. Pain and its treatment in outpatients with metastatic cancer. N Engl J Med 1994; 330: 592–596. 7. Costantini M, Viterboni P, Flego G. Prevalence of pain in Italian hospitals: results of a regional cross-sectional survey. J Pain Symptom Manage 2002; 23: 221–230. 8. Ripamonti C, Zecca E, Brunelli C et al. Pain experienced by patients hospitalized at the national Cancer Institute of Milan: research project ‘Towards a pain-free hospital’. Tumori 2000; 86: 412–418. 9. Visentin M, Zanolin E, Trentin L et al. Prevalence and treatment of pain in adults admitted to Italian hospital. Eur J Pain 2005; 9: 61–67. 10. Sichetti D, Di Biagio K, Anino N et al. Il Modello di sorveglianza ECADO—Epidemiologia Clinico-Assistenziale del Dolore in Osdpedale: il contributo degli infermieri. Assistenza infermieristica e ricerca 2009; 28(2): 73–81. 11. World Health Organization. Cancer Pain Relief. 2nd edition. Geneva, Switzerland: World Health Organization, 1996. 12. Holtan A, Aass N, Nordøy T et al. Prevalence of pain in hospitalised cancer patients in Norway: a national survey. Palliat Med 2007; 21: 7–13. 13. Foley KM, Kathleen M. The treatment of cancer pain. New Engl J Med 1985; 313: 84–95. 14. Morselli M, Bandieri E, Zanin R et al. Pain and emotional distress in leukemia patients at diagnosis. Leuk Res 2010; 34: e67–e68. 15. Zenz M, Willweber-Strumpf A. Opiophobia and cancer pain in Europe. Lancet 1993; 341: 1075–1076. 16. Paice JA, Toy C, Shott S. Barriers to cancer pain relief: fear of tolerance and addiction. J Pain Symptom Manage 1998; 16: 1–9. 17. Weiss SC, Emanuel LL, Fairclough DL, Emanuel EJ. Understanding the experience of pain in terminally ill patients. Lancet 2001; 357: 1311–1315. 18. Ward SE, Goldberg N, Miller-McCauley M et al. Patient-related barriers to management of cancer pain. Pain 1993; 52: 319–324. 19. Bandieri E, Chirarolanza A, Luppi M et al. Prescription of opioids in Italy: everything but the morphine [letters to the editor]. Ann Oncol 2009; 20: 961–962.
doi:10.1093/annonc/mdq155 | 2093
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 19, 2015
ASL 16 Piemonte: A. Bramardi, B. Bovetti, L. Bagnasco; Monselice (PD), PO ULSS 17 di Este: R. Congedo, A. Amato, M. Cogo, F. Visentini, G. Reffo, F. Bellamio, N. Marconcini; Montebelluna (TV), Ospedali Azienda ULSS 3 Veneto: R. Callegari, M. Osti, M. Ragazzi, G. Cecchetto, L. Tessaro, L. Lusiani, S. Barichello, F. Sartor, M. Cervi, G. Rech, F. Patassini, M. Gottardo; Montecchio Emilia (RE), Osp. E. Franchini: S. Fietta, G. Pagliarini, A. Cocconi, R. Aldini; Palermo, AO Civico: P. Finocchiaro, P. Marrone, G. Trizzino, M. Di Liberto, M. Tamburo De Bella; Pescara, Osp. Civile S. Spirito: F. Sanita`, M. Di Staso, M. Lombardo, G. Gidaro, P. Lelli Chiesa, G. Parruti, L. Ceccomancini, I. Pelusi; Pisa, AOU Pisana: M. Polvani, F. Vivaldi, A. Paolicchi, M. Marcaccini, C. Venturi; Ragusa, AO Civile Maria Paterno` Arezzo: M. Martorana; Roma, AOU Policlinico Tor Vergata: M. G. Celeste, A. Marra, L. Pacca, G. Visconti, A. Gatti, C. Cappitella; Rovereto (TN), Osp. S. Maria del Carmine: G. Dusi, M. Mattarei, R. Girardello, R. Falzone, N. Stefenelli, M. Saiani, S. Scarabello Vettore; Senigallia (AN), Osp. di Senigallia: P. Borona, R. Connestari, V. M. Carfi; Taranto, AO SS. Annunziata: R. Moscogiuri, C. A. Pennetta, D. Zollino, D. Semeraro, G. Caldaralo, M. A. De Maria; Teramo, Osp. Mazzini: M. Turchetti, I. Senesi, O. Cipolletti, S. Erpini, P. Lisciani; Tradate (VA), Osp. Galmarini: P. Lusuriello, A. Volonte`, M. D. Centemeri; Trento, Osp. di Trento: M. Boni, A. Campomori, F. Branz, E. Geat, M. Brugnaria, E. Modena; Vasto (CH), Osp. S. Pio Da Pietralcina: L. Di Fabio, C. Spoltore, E. Testa; Vercelli, Osp. Sant’Andrea: M. Giolito, M. Cianfanelli.
original article