ORIGINAL PAPER
Pain management in Internal Medicine and effects of a standardised educational intervention: the FADOIDOMINO Study G. Civardi,1 F. Zucco,2 A. Valerio,3 A. Fontanella,4 M. Alessandri,5 C. D. Ciannamea,6 L. Perale,7 G. Gussoni,3 E. Bonizzoni,8 G. Vescovo,9 G. Agnelli,10 M. Campanini,11 A. Mazzone,12 C. Nozzoli,13 for the Research Department of FADOI*
1
SUMMARY
What’s known
Purpose: Few real-world data are available on the frequency and management of pain in Internal Medicine (IM). Aims of our study were to assess the prevalence of pain in IM, and to evaluate the effects on pain management of a standardised educational programme. Materials and methods: The study was performed in 26 IM Units in Italy, with two cross-sectional surveys (PRE phase and POST phase) interspersed with an educational programme. In PRE phase each Centre reviewed the hospital charts of the last 100 consecutive patients hospitalised for any cause. An educational programme was conducted in each Centre by means of the ‘outreach visit’, a face-to-face meeting between health personnel and a trained external expert. Six months after, each Centre repeated the data collection (POST phase), specular to the PRE. Results: A total of 5200 medical charts were analysed. Pain was documented in 37.5% of the patients. After the educational intervention, the intensity of pain was appropriately assessed in a higher percentage of patients (77.4% vs. 47.8%, p = 0.0001), and it was more frequently monitored during hospitalisation. Qualitative definition of pain (pathogenesis, duration, etc.) increased in POST phase (75.4% vs. 62.7%, p = 0.0001). A 73.3% increase in the use of strong opioids was detected following educational programme. Conclusions: Pain affects 4 out of 10 patients hospitalised in IM. According to our large real-world study, to implement a standardised one-shot educational programme may persistently improve the attitude of health personnel towards the characterisation and management of pain.
Introduction In spite of increasing attention towards the problem of classification and treatment of pain in hospitalised patients, and the availability of a number of epidemiological studies (1–3), both prevalence and management of this symptom are still poorly defined, namely within the setting of Internal Medicine (IM). Patients hospitalised in IM wards are heterogeneous, frequently elderly and frail, with a relevant burden of comorbidity and multiple therapies: these features make challenging a precise assessment of the prevalence of pain and its optimal treatment in hospitalised medical patients. ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 33–40. doi: 10.1111/ijcp.12515
Pain is a not negligible finding among patients hospitalised in Internal Medicine, but epidemiological data in this setting are poor. In the last years, the attention towards the problem of pain management in Italian hospitals has grown also because of a specific national law stating that pain has to be registered in hospital charts, and its intensity monitored. Information on the adherence to these recommendations are however lacking. Furthermore, the use of strong opioid in Italy is traditionally low.
What’s new In this multicenter study, pain appeared as a frequent issue, affecting 4 out of 10 hospitalised patients. Significant changes in the in-hospital management of patients with pain (namely, increase in the percentage of patients with measurement and monitoring of the intensity of pain, and with appropriate assessment of pain pathophysiology) have been detected following a systematic educational intervention. Furthermore, a significantly higher percentage of patients receiving specific treatment for pain and particularly strong opioids, were observed posttraining.
In the last 10 years, the attention to the problem of pain management in Italian hospitals has grown exponentially. Some well-conducted epidemiological studies (4,5) have reported high prevalences of pain among patients admitted to Italian hospitals, and shown that the treatments used were not sufficient to control the symptom (6). In May 2001, the Italian government promoted the project ‘Pain–free Hospital’ that involves all hospitals in the Italian territory, with the aim to improve clinical care strategies for pain of any origin. In this programme, attention was mainly focused on the issues of postoperative pain, cancer pain and pain from childbirth, while remaining in the backstage the problem of pain control in
Internal Medicine, Hospital of Fiorenzuola d’Arda, Piacenza, Italy 2 Department of Anesthesiology, Palliative Care and Therapy of Pain, Hospital G. Salvini, Garbagnate M.se, Milan, Italy 3 Research Department, FADOI Foundation, Milan, Italy 4 Internal Medicine, Hospital Fatebenefratelli, Naples, Italy 5 Internal Medicine, Misericordia Hospital, Grosseto, Italy 6 Internal Medicine, Hospital of Molfetta, Bari, Italy 7 Internal Medicine, Hospital S. Antonio, San Daniele del Friuli, Udine, Italy 8 Department of Clinical Science and Community, Section of Medical Statistics and Biometry “GA Maccacaro”, University of Milan, Milan, Italy 9 Internal Medicine, Hospital S. Bortolo, Vicenza, Italy 10 Internal and Cardiovascular Medicine – Stroke Unit, University of Perugia, Perugia, Italy 11 Department of Internal Medicine, Hospital Maggiore della Carita, Novara, Italy 12 Internal Medicine, Hospital Civile, Legnano, Italy 13 Internal Medicine, Hospital Careggi, Florence, Italy Correspondence to: Gualberto Gussoni, MD, Clinical Research Department, FADOI Foundation, Piazzale L. Cadorna, 15 – 20123 – Milan, Italy Tel.: + 39 02 48005140 Fax: + 39 06 85345986 Email: gualberto.gussoni@ gmail.com *Members of the FADOIDOMINO Study Group are listed in the Appendix at the end of paper. FADOI is the Italian
33
34
Pain management in Internal Medicine
Scientific Society of Hospital Internal Medicine.
medical patients. Furthermore, a national law was released in 2010 stating that pain has to be registered in hospital charts, and its intensity monitored just like vital parameters such as temperature, heart rate and blood pressure (7). A preliminary survey conducted by the Italian Federation of Hospital Internists (FADOI) in March 2010, which involved 124 units of IM, has shown that more than half of these units had not formally adopted an internal procedure for the optimal treatment nor for the measurement of pain. However, 68% of physicians considered pain as a major clinical problem: moreover, 10% among these looked at this problem as a true priority for hospitalised medical patients (data on file, Clinical Research Department, FADOI Foundation). These data confirm the existence of dissociation between perception of the problem and its systematic detection and management. Based on these findings, aim of the DOMINO (DOlore in Medicina Interna NO = No pain in IM) study carried out by the Italian Scientific Society FADOI was to better define the prevalence and the attitude towards management of pain in patients admitted to Departments of Internal Medicine. In addition, the possible effects of a standardised educational programme, mainly focused on the evaluation and characterisation of pain, and general treatment strategies, were evaluated.
Disclosures All the authors have declared no conflicts of interest.
Materials and methods The FADOI-DOMINO study has been performed from January to December 2011 in a pool of 26 Italian Centres of Internal Medicine considered ‘sensitive to the theme of pain management’ and representative of the national health system (as for geographical distribution, size of the hospital, level of intensity of care by the Unit). The Centres were selected among those having someone within health personnel who had previously attended educational events on this topic promoted by the Scientific Society FADOI. A multidisciplinary Scientific Committee, including experts in IM and Pain therapy, was responsible for study planning and conduction. The study design included two cross-sectional surveys (PRE phase and POST phase) interspersed with an educational programme. The FADOI-DOMINO study was conducted in accordance with the existing national rules. The study protocol received formal approval of the Ethical Committee of the Coordinating Centre, and it was notified to the Ethical Committees of all the other participating Centres.
PRE phase Each of the 26 participating Centres reviewed the hospital charts of the last 100 consecutive patients hospitalised for any cause for at least 24 h, starting from a date prior to the involvement of the Centre in the project. At the time of the study, around 30% of participating centres had internal procedures for the management of pain, however not standardised. General information (age, gender, active chronic or acute diseases, conditions related to the discharge) were collected for each patient. Furthermore, an accurate review of hospital charts and nursing care diary was done to select those patients for which citation of pain, or pain measurement or specific treatment for pain were present (‘pain group’). In this subgroup of patients additional information, if available, and specifically relevant to pain were retrieved from clinical documentation: • classification (aetiology/physiopathology – sites – duration); • measurement of intensity (if any: type of method used for assessment, number of detections during hospital stay, professional who carried out the survey); • treatment for pain, during hospitalisation and at discharge (pharmacological, invasive, psychological); • supportive therapies (antidepressants, antiepilectics, corticosteroids, laxatives, anti-emetics, etc.); • side effects related to pain therapy; • address to pain specialists and/or pain clinics during hospitalisation and at discharge.
Educational programme Following the first data collection, a standardised educational intervention was performed by using the technique of the so called ‘outreach visits’. In each participating Centre, a face-to-face meeting was scheduled between an expert teacher external to the Unit and the staff (physicians and nurses) of the Internal Medicine Centre. Contents of the educational intervention had been previously defined and standardised accordingly to international guidelines, as well as by considering major deviations from best medical practice detected in PRE phase. A standardised slide-kit was used for presentation during the outreach visits, and it remained at disposal of the health personnel of participating Centres. The main topics of the slide-kit were the following: definitions, classification, aetiology and pathophysiology of pain; pain in cancer and non-cancer patients (with particular reference to elderly); assessment, measurement and monitoring of pain; basic update for pain therapy. Prior to execution of the outreach visits, the teachers received an adequate training coordinated ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 33–40
Pain management in Internal Medicine
by the Scientific Board. The technique of the outreach visits was selected since it was reported as effective in improving professional practice and healthcare outcomes (8). Globally speaking, 338 physicians or nurses attended the outreach visits in the 26 Centres (mean 13, range 9–19).
POST phase Around 6 months after the ‘outreach visit’ (to evaluate persistence of effect of the educational intervention), each Centre repeated the data collection in a specular way as for PRE phase. It was therefore possible to compare, in homogeneous experimental conditions, the two settings of data obtained in the pre-educational vs. the post-educational phase of the study. The following items were selected a priori as indicators of the effects of training: • Percentage of patients with an assessment of pain pathophysiology. • Percentage of patients who received measurement of intensity of pain by means of validated methods. • Percentage of patients who had measurement of intensity of pain more than once during hospital stay. • Percentage of patients receiving specific treatment within the pain group. • Percentage of patients treated with strong opioids. • Percentage of patients treated with specific adjuvant therapies for pain (namely, antiepileptic and antidepressant drugs). • Percentage of patients undergoing clinical evaluation by Pain or Palliative Care specialist during hospitalisation, or addressed to pain specialists and/or pain clinics at discharge.
Statistical analysis Descriptive statistics refer to all included patients. For continuous variables, the mean and standard deviation were calculated. For each discrete variable, the percentage (prevalence) was calculated as the number of cases in each category, in relation to all cases with non-missing values of that variable. In general, statistical comparisons were performed using one-way ANOVA models for the analyses carried out on continuous variables and with the v2 test or Fisher’s exact test, if deemed more appropriate, for the analyses carried out on categorical variables. Since no previous data were available in the literature on the objectives evaluated in the present investigation, and taking into account of the exploratory nature of the study, no hypothesis of results could be done. In consequences, no sample size estimation has been calculated and 100 patients for each Centre were ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 33–40
planned to be enrolled. This figure represented the maximum possible effort from the economical and feasibility viewpoint. All statistical analyses were generated with the use of standard procedures of SAS software, version 9.2 (SAS Institute, Cary, NC).
Results A total of 5200 medical charts were analysed, 2600 in PRE phase and 2600 in POST phase, respectively. The general characteristics of study population in PRE phase and in POST phase were very similar (Table 1); confirming the frequent high level of comorbidity of patients hospitalised in IM, 61.3% of patients had at least three concomitant active diseases. As a first result emerging from this study, 35.1% and 37.5% of patients (p = 0.07) were classified within the ‘pain group’ (i.e. patients with explicit citation of the symptom in hospital documentation, measurement of intensity or specific treatment for pain) in PRE and POST phase, respectively. After the educational intervention, a higher percentage of patients within the pain group had an appropriate qualitative and quantitative definition of pain (pathogenesis, duration) (Figure 1). The percentages of patients with cancer or non-cancer pain (considering both PRE- and POST phase) were 23.2% and 76.8%, respectively; in the subgroup with non-cancer pain, more than half of patients had osteo-articular/muscular pain (Table 2). Assessment of intensity of pain, by means of validated methods, was done in significantly more patients in POST if compared with PRE phase (77.4% vs. 47.8%, p = 0.0001); an increase was also observed in the percentage of patients who had multiple measurements during hospital stay (Figure 2). The Numeric Rating Scale (0–10) was the most frequently used method for evaluation of intensity of pain (in around 60% of patients), followed by Verbal Assessment Scale and Visual Analogic Scale. Measurement and registration of pain intensity were performed by physicians only in 19.5% of patients, by nurses only in 50.7% and by both physicians and nurses in 29.8% of cases, without significant differences between the two study phases. A significantly higher percentage of patients hospitalised in IM in POST phase received specific treatment for pain, if compared with PRE phase (32.4% vs. 28.3%, p = 0.001). The details of therapies are described in Figure 3 and Table 3. To note, a 73.3% increase in the use of ‘strong’ opioids was detected at POST phase, if compared with PRE phase. Adverse events considered related to the treatment of pain occurred in 12.4% and 11.7% of patients of pain group in PRE and POST phases (p = 0.70), being
35
36
Pain management in Internal Medicine
Table 1 General characteristics of patients evaluated in PRE and POST phases
Age (years) < 50 50–69 70–89 ≥ 90 Female gender Language or cognitive impairment Anxiety or depression Cancer Outcome of hospitalisation Discharge Transfer to other Units Death
PRE phase (n = 2600)
POST phase (n = 2600)
p-value
8.7 20.5 63.1 7.7 51.4 18.2 24.6 18.9
8.8 20.1 62.4 8.2 50.7 17.5 22.5 20.3
0.84
81.7 8.4 9.9
84.2 7.0 8.8
0.05
0.62 0.49 0.07 0.21
Values are expressed in percentages.
Table 2 Pathogenesis of pain (PRE + POST phase)
Pathogenesis of pain
% (n = 1577)
Cancer pain Non-cancer pain Osteo-articular-muscle Visceral abdominal Neuropathic Cardiac Visceral pelvic Mixed Not classified
23.2 76.8 41.3 15.2 6.0 3.1 1.0 3.9 6.5
Figure 1 Percentage of patients with a qualitative and/or quantitative classification of pain within the ‘pain group’
constipation the most frequent (19.0% and 12.4% among patients receiving strong opioids in PRE and POST phase, respectively, p = 0.03). In the total study population, adjuvant therapy for pain was used in 12.4% and 13.7% of patients (p = 0.15). Corticosteroids were the most frequently
used drugs (4.3% and 5.5% in PRE and POST, p = 0.05), treatment with antidepressants as supportive therapy for pain was carried out in 3.0% and 4.1% (p = 0.03), anti-epileptics were used in 1.1% and 1.7% (p = 0.06) in the two study phases, respectively. Table 4 summarises the findings of the study as for address to pain specialists and/or pain clinics or palliative care during hospitalisation and at discharge, for patients of the ‘pain group’.
Discussion To our knowledge, this is the first study that evaluates in details the prevalence and characteristics of pain in a population of patients admitted to IM Departments. In this multicenter study, pain appeared as a frequent issue, affecting 4 out of 10 hospitalised patients. Significant changes in the inhospital management of patients with pain (namely, increase in the percentage of patients with measurement and monitoring of the intensity of pain, and with appropriate assessment of pain pathophysiology) have been detected following a systematic educational intervention. Furthermore, a significantly higher percentage of patients receiving specific treatment for pain, and particularly strong opioids, were observed posttraining. This finding is of particular interest, since the strong opioid consumption in Italy is traditionally low (9). Some authors (4,5) have previously investigated the prevalence of pain in Italian hospitals and the partial data obtained from medical wards substantially confirm the prevalence reported in our study. Other authors reported a higher prevalence by ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 33–40
Pain management in Internal Medicine
Figure 2 Percentage of patients with at least one measurement of pain intensity (left panel) within ‘pain group’, and frequency of monitoring of pain intensity during hospital stay (right panel)
Table 3 Drugs used for treatment of pain in PRE- and POST phases
Paracetamol only NSAIDs only Weak opioids only Combined therapy without strong opiods Strong opiods only Combined therapy with strong opiods Strong opioids total
PRE phase (n = 2600)
POST phase (n = 2600)
167 86 119 150 98 112 210
169 73 210 33 276 88 364
(6.4) (3.3) (4.6) (5.8) (3.8) (4.3) (8.1)
(6.5) (2.8) (8.0) (1.2) (10.6) (3.4) (14.0)
p-value
< < <
Pain management in Internal Medicine and effects of a standardised educational intervention: the FADOIDOMINO Study G. Civardi,1 F. Zucco,2 A. Valerio,3 A. Fontanella,4 M. Alessandri,5 C. D. Ciannamea,6 L. Perale,7 G. Gussoni,3 E. Bonizzoni,8 G. Vescovo,9 G. Agnelli,10 M. Campanini,11 A. Mazzone,12 C. Nozzoli,13 for the Research Department of FADOI*
1
SUMMARY
What’s known
Purpose: Few real-world data are available on the frequency and management of pain in Internal Medicine (IM). Aims of our study were to assess the prevalence of pain in IM, and to evaluate the effects on pain management of a standardised educational programme. Materials and methods: The study was performed in 26 IM Units in Italy, with two cross-sectional surveys (PRE phase and POST phase) interspersed with an educational programme. In PRE phase each Centre reviewed the hospital charts of the last 100 consecutive patients hospitalised for any cause. An educational programme was conducted in each Centre by means of the ‘outreach visit’, a face-to-face meeting between health personnel and a trained external expert. Six months after, each Centre repeated the data collection (POST phase), specular to the PRE. Results: A total of 5200 medical charts were analysed. Pain was documented in 37.5% of the patients. After the educational intervention, the intensity of pain was appropriately assessed in a higher percentage of patients (77.4% vs. 47.8%, p = 0.0001), and it was more frequently monitored during hospitalisation. Qualitative definition of pain (pathogenesis, duration, etc.) increased in POST phase (75.4% vs. 62.7%, p = 0.0001). A 73.3% increase in the use of strong opioids was detected following educational programme. Conclusions: Pain affects 4 out of 10 patients hospitalised in IM. According to our large real-world study, to implement a standardised one-shot educational programme may persistently improve the attitude of health personnel towards the characterisation and management of pain.
Introduction In spite of increasing attention towards the problem of classification and treatment of pain in hospitalised patients, and the availability of a number of epidemiological studies (1–3), both prevalence and management of this symptom are still poorly defined, namely within the setting of Internal Medicine (IM). Patients hospitalised in IM wards are heterogeneous, frequently elderly and frail, with a relevant burden of comorbidity and multiple therapies: these features make challenging a precise assessment of the prevalence of pain and its optimal treatment in hospitalised medical patients. ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 33–40. doi: 10.1111/ijcp.12515
Pain is a not negligible finding among patients hospitalised in Internal Medicine, but epidemiological data in this setting are poor. In the last years, the attention towards the problem of pain management in Italian hospitals has grown also because of a specific national law stating that pain has to be registered in hospital charts, and its intensity monitored. Information on the adherence to these recommendations are however lacking. Furthermore, the use of strong opioid in Italy is traditionally low.
What’s new In this multicenter study, pain appeared as a frequent issue, affecting 4 out of 10 hospitalised patients. Significant changes in the in-hospital management of patients with pain (namely, increase in the percentage of patients with measurement and monitoring of the intensity of pain, and with appropriate assessment of pain pathophysiology) have been detected following a systematic educational intervention. Furthermore, a significantly higher percentage of patients receiving specific treatment for pain and particularly strong opioids, were observed posttraining.
In the last 10 years, the attention to the problem of pain management in Italian hospitals has grown exponentially. Some well-conducted epidemiological studies (4,5) have reported high prevalences of pain among patients admitted to Italian hospitals, and shown that the treatments used were not sufficient to control the symptom (6). In May 2001, the Italian government promoted the project ‘Pain–free Hospital’ that involves all hospitals in the Italian territory, with the aim to improve clinical care strategies for pain of any origin. In this programme, attention was mainly focused on the issues of postoperative pain, cancer pain and pain from childbirth, while remaining in the backstage the problem of pain control in
Internal Medicine, Hospital of Fiorenzuola d’Arda, Piacenza, Italy 2 Department of Anesthesiology, Palliative Care and Therapy of Pain, Hospital G. Salvini, Garbagnate M.se, Milan, Italy 3 Research Department, FADOI Foundation, Milan, Italy 4 Internal Medicine, Hospital Fatebenefratelli, Naples, Italy 5 Internal Medicine, Misericordia Hospital, Grosseto, Italy 6 Internal Medicine, Hospital of Molfetta, Bari, Italy 7 Internal Medicine, Hospital S. Antonio, San Daniele del Friuli, Udine, Italy 8 Department of Clinical Science and Community, Section of Medical Statistics and Biometry “GA Maccacaro”, University of Milan, Milan, Italy 9 Internal Medicine, Hospital S. Bortolo, Vicenza, Italy 10 Internal and Cardiovascular Medicine – Stroke Unit, University of Perugia, Perugia, Italy 11 Department of Internal Medicine, Hospital Maggiore della Carita, Novara, Italy 12 Internal Medicine, Hospital Civile, Legnano, Italy 13 Internal Medicine, Hospital Careggi, Florence, Italy Correspondence to: Gualberto Gussoni, MD, Clinical Research Department, FADOI Foundation, Piazzale L. Cadorna, 15 – 20123 – Milan, Italy Tel.: + 39 02 48005140 Fax: + 39 06 85345986 Email: gualberto.gussoni@ gmail.com *Members of the FADOIDOMINO Study Group are listed in the Appendix at the end of paper. FADOI is the Italian
33
34
Pain management in Internal Medicine
Scientific Society of Hospital Internal Medicine.
medical patients. Furthermore, a national law was released in 2010 stating that pain has to be registered in hospital charts, and its intensity monitored just like vital parameters such as temperature, heart rate and blood pressure (7). A preliminary survey conducted by the Italian Federation of Hospital Internists (FADOI) in March 2010, which involved 124 units of IM, has shown that more than half of these units had not formally adopted an internal procedure for the optimal treatment nor for the measurement of pain. However, 68% of physicians considered pain as a major clinical problem: moreover, 10% among these looked at this problem as a true priority for hospitalised medical patients (data on file, Clinical Research Department, FADOI Foundation). These data confirm the existence of dissociation between perception of the problem and its systematic detection and management. Based on these findings, aim of the DOMINO (DOlore in Medicina Interna NO = No pain in IM) study carried out by the Italian Scientific Society FADOI was to better define the prevalence and the attitude towards management of pain in patients admitted to Departments of Internal Medicine. In addition, the possible effects of a standardised educational programme, mainly focused on the evaluation and characterisation of pain, and general treatment strategies, were evaluated.
Disclosures All the authors have declared no conflicts of interest.
Materials and methods The FADOI-DOMINO study has been performed from January to December 2011 in a pool of 26 Italian Centres of Internal Medicine considered ‘sensitive to the theme of pain management’ and representative of the national health system (as for geographical distribution, size of the hospital, level of intensity of care by the Unit). The Centres were selected among those having someone within health personnel who had previously attended educational events on this topic promoted by the Scientific Society FADOI. A multidisciplinary Scientific Committee, including experts in IM and Pain therapy, was responsible for study planning and conduction. The study design included two cross-sectional surveys (PRE phase and POST phase) interspersed with an educational programme. The FADOI-DOMINO study was conducted in accordance with the existing national rules. The study protocol received formal approval of the Ethical Committee of the Coordinating Centre, and it was notified to the Ethical Committees of all the other participating Centres.
PRE phase Each of the 26 participating Centres reviewed the hospital charts of the last 100 consecutive patients hospitalised for any cause for at least 24 h, starting from a date prior to the involvement of the Centre in the project. At the time of the study, around 30% of participating centres had internal procedures for the management of pain, however not standardised. General information (age, gender, active chronic or acute diseases, conditions related to the discharge) were collected for each patient. Furthermore, an accurate review of hospital charts and nursing care diary was done to select those patients for which citation of pain, or pain measurement or specific treatment for pain were present (‘pain group’). In this subgroup of patients additional information, if available, and specifically relevant to pain were retrieved from clinical documentation: • classification (aetiology/physiopathology – sites – duration); • measurement of intensity (if any: type of method used for assessment, number of detections during hospital stay, professional who carried out the survey); • treatment for pain, during hospitalisation and at discharge (pharmacological, invasive, psychological); • supportive therapies (antidepressants, antiepilectics, corticosteroids, laxatives, anti-emetics, etc.); • side effects related to pain therapy; • address to pain specialists and/or pain clinics during hospitalisation and at discharge.
Educational programme Following the first data collection, a standardised educational intervention was performed by using the technique of the so called ‘outreach visits’. In each participating Centre, a face-to-face meeting was scheduled between an expert teacher external to the Unit and the staff (physicians and nurses) of the Internal Medicine Centre. Contents of the educational intervention had been previously defined and standardised accordingly to international guidelines, as well as by considering major deviations from best medical practice detected in PRE phase. A standardised slide-kit was used for presentation during the outreach visits, and it remained at disposal of the health personnel of participating Centres. The main topics of the slide-kit were the following: definitions, classification, aetiology and pathophysiology of pain; pain in cancer and non-cancer patients (with particular reference to elderly); assessment, measurement and monitoring of pain; basic update for pain therapy. Prior to execution of the outreach visits, the teachers received an adequate training coordinated ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 33–40
Pain management in Internal Medicine
by the Scientific Board. The technique of the outreach visits was selected since it was reported as effective in improving professional practice and healthcare outcomes (8). Globally speaking, 338 physicians or nurses attended the outreach visits in the 26 Centres (mean 13, range 9–19).
POST phase Around 6 months after the ‘outreach visit’ (to evaluate persistence of effect of the educational intervention), each Centre repeated the data collection in a specular way as for PRE phase. It was therefore possible to compare, in homogeneous experimental conditions, the two settings of data obtained in the pre-educational vs. the post-educational phase of the study. The following items were selected a priori as indicators of the effects of training: • Percentage of patients with an assessment of pain pathophysiology. • Percentage of patients who received measurement of intensity of pain by means of validated methods. • Percentage of patients who had measurement of intensity of pain more than once during hospital stay. • Percentage of patients receiving specific treatment within the pain group. • Percentage of patients treated with strong opioids. • Percentage of patients treated with specific adjuvant therapies for pain (namely, antiepileptic and antidepressant drugs). • Percentage of patients undergoing clinical evaluation by Pain or Palliative Care specialist during hospitalisation, or addressed to pain specialists and/or pain clinics at discharge.
Statistical analysis Descriptive statistics refer to all included patients. For continuous variables, the mean and standard deviation were calculated. For each discrete variable, the percentage (prevalence) was calculated as the number of cases in each category, in relation to all cases with non-missing values of that variable. In general, statistical comparisons were performed using one-way ANOVA models for the analyses carried out on continuous variables and with the v2 test or Fisher’s exact test, if deemed more appropriate, for the analyses carried out on categorical variables. Since no previous data were available in the literature on the objectives evaluated in the present investigation, and taking into account of the exploratory nature of the study, no hypothesis of results could be done. In consequences, no sample size estimation has been calculated and 100 patients for each Centre were ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 33–40
planned to be enrolled. This figure represented the maximum possible effort from the economical and feasibility viewpoint. All statistical analyses were generated with the use of standard procedures of SAS software, version 9.2 (SAS Institute, Cary, NC).
Results A total of 5200 medical charts were analysed, 2600 in PRE phase and 2600 in POST phase, respectively. The general characteristics of study population in PRE phase and in POST phase were very similar (Table 1); confirming the frequent high level of comorbidity of patients hospitalised in IM, 61.3% of patients had at least three concomitant active diseases. As a first result emerging from this study, 35.1% and 37.5% of patients (p = 0.07) were classified within the ‘pain group’ (i.e. patients with explicit citation of the symptom in hospital documentation, measurement of intensity or specific treatment for pain) in PRE and POST phase, respectively. After the educational intervention, a higher percentage of patients within the pain group had an appropriate qualitative and quantitative definition of pain (pathogenesis, duration) (Figure 1). The percentages of patients with cancer or non-cancer pain (considering both PRE- and POST phase) were 23.2% and 76.8%, respectively; in the subgroup with non-cancer pain, more than half of patients had osteo-articular/muscular pain (Table 2). Assessment of intensity of pain, by means of validated methods, was done in significantly more patients in POST if compared with PRE phase (77.4% vs. 47.8%, p = 0.0001); an increase was also observed in the percentage of patients who had multiple measurements during hospital stay (Figure 2). The Numeric Rating Scale (0–10) was the most frequently used method for evaluation of intensity of pain (in around 60% of patients), followed by Verbal Assessment Scale and Visual Analogic Scale. Measurement and registration of pain intensity were performed by physicians only in 19.5% of patients, by nurses only in 50.7% and by both physicians and nurses in 29.8% of cases, without significant differences between the two study phases. A significantly higher percentage of patients hospitalised in IM in POST phase received specific treatment for pain, if compared with PRE phase (32.4% vs. 28.3%, p = 0.001). The details of therapies are described in Figure 3 and Table 3. To note, a 73.3% increase in the use of ‘strong’ opioids was detected at POST phase, if compared with PRE phase. Adverse events considered related to the treatment of pain occurred in 12.4% and 11.7% of patients of pain group in PRE and POST phases (p = 0.70), being
35
36
Pain management in Internal Medicine
Table 1 General characteristics of patients evaluated in PRE and POST phases
Age (years) < 50 50–69 70–89 ≥ 90 Female gender Language or cognitive impairment Anxiety or depression Cancer Outcome of hospitalisation Discharge Transfer to other Units Death
PRE phase (n = 2600)
POST phase (n = 2600)
p-value
8.7 20.5 63.1 7.7 51.4 18.2 24.6 18.9
8.8 20.1 62.4 8.2 50.7 17.5 22.5 20.3
0.84
81.7 8.4 9.9
84.2 7.0 8.8
0.05
0.62 0.49 0.07 0.21
Values are expressed in percentages.
Table 2 Pathogenesis of pain (PRE + POST phase)
Pathogenesis of pain
% (n = 1577)
Cancer pain Non-cancer pain Osteo-articular-muscle Visceral abdominal Neuropathic Cardiac Visceral pelvic Mixed Not classified
23.2 76.8 41.3 15.2 6.0 3.1 1.0 3.9 6.5
Figure 1 Percentage of patients with a qualitative and/or quantitative classification of pain within the ‘pain group’
constipation the most frequent (19.0% and 12.4% among patients receiving strong opioids in PRE and POST phase, respectively, p = 0.03). In the total study population, adjuvant therapy for pain was used in 12.4% and 13.7% of patients (p = 0.15). Corticosteroids were the most frequently
used drugs (4.3% and 5.5% in PRE and POST, p = 0.05), treatment with antidepressants as supportive therapy for pain was carried out in 3.0% and 4.1% (p = 0.03), anti-epileptics were used in 1.1% and 1.7% (p = 0.06) in the two study phases, respectively. Table 4 summarises the findings of the study as for address to pain specialists and/or pain clinics or palliative care during hospitalisation and at discharge, for patients of the ‘pain group’.
Discussion To our knowledge, this is the first study that evaluates in details the prevalence and characteristics of pain in a population of patients admitted to IM Departments. In this multicenter study, pain appeared as a frequent issue, affecting 4 out of 10 hospitalised patients. Significant changes in the inhospital management of patients with pain (namely, increase in the percentage of patients with measurement and monitoring of the intensity of pain, and with appropriate assessment of pain pathophysiology) have been detected following a systematic educational intervention. Furthermore, a significantly higher percentage of patients receiving specific treatment for pain, and particularly strong opioids, were observed posttraining. This finding is of particular interest, since the strong opioid consumption in Italy is traditionally low (9). Some authors (4,5) have previously investigated the prevalence of pain in Italian hospitals and the partial data obtained from medical wards substantially confirm the prevalence reported in our study. Other authors reported a higher prevalence by ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 33–40
Pain management in Internal Medicine
Figure 2 Percentage of patients with at least one measurement of pain intensity (left panel) within ‘pain group’, and frequency of monitoring of pain intensity during hospital stay (right panel)
Table 3 Drugs used for treatment of pain in PRE- and POST phases
Paracetamol only NSAIDs only Weak opioids only Combined therapy without strong opiods Strong opiods only Combined therapy with strong opiods Strong opioids total
PRE phase (n = 2600)
POST phase (n = 2600)
167 86 119 150 98 112 210
169 73 210 33 276 88 364
(6.4) (3.3) (4.6) (5.8) (3.8) (4.3) (8.1)
(6.5) (2.8) (8.0) (1.2) (10.6) (3.4) (14.0)
p-value
< < <
