Current Medical Research & Opinion 0300-7995 doi:10.1185/03007995.2014.884492
Vol. 30, No. 6, 2014, 1033–1042
Article ST-0372.R1/884492 All rights reserved: reproduction in whole or part not permitted
Curr Med Res Opin Downloaded from informahealthcare.com by Nyu Medical Center on 02/19/15 For personal use only.
Original article COPD management according to old and new GOLD guidelines: an observational study with Italian general practitioners
S. Maio S. Baldacci F. Martini S. Cerrai G. Sarno M. Borbotti Pulmonary Environmental Epidemiology Unit, CNR Institute of Clinical Physiology, Pisa, Italy
Abstract Background: Guideline recommendations for COPD management are only partially applied within primary care clinical practice. Objective: To compare the COPD management by Italian general practitioners (GPs) according to either the old GOLD (oGOLD) or the new GOLD (nGOLD) guidelines.
A.P. Pala Technosciences Unit, CNR Institute of Clinical Physiology, Pisa, Italy
N. Murgia Section of Occupational Medicine, Respiratory Diseases and Toxicology, University of Perugia, Perugia, Italy
G. Viegi Pulmonary Environmental Epidemiology Unit, CNR Institute of Clinical Physiology, Pisa, Italy
on behalf of the COMODHES study group Address for correspondence: Dr. Sara Maio, CNR Institute of Clinical Physiology, Via Trieste, 41 - 56126 Pisa, Italy. Tel: +39 050-502031; Fax: +39 050-503596; [email protected] Keywords: COPD – General practitioners – GOLD guidelines – Prescriptions pattern Accepted: 6 January 2014; published online: 3 February 2014 Citation: Curr Med Res Opin 2014; 30:1033–42
Research design and methods: Observational study in different Italian areas. A total of 176 GPs enrolled their patients with a COPD diagnosis. Questionnaires were used to collect data on: COPD symptoms, disease severity, exacerbations, prescribed pharmacological and non-pharmacological treatments. COPD severity was estimated according to oGOLD and nGOLD guidelines. Results: A total of 526 subjects had complete information to assess COPD severity level according to guidelines (symptoms level, spirometry, history of exacerbations). The investigated subjects were more frequently males (71.2%) with a mean age of 72.5 years, and ex-smokers (44.4%). GPs reported sufficient control of the disease in 47.2% of the subjects with over two exacerbations in the last 12 months. Most patients have moderate COPD (51.5%), according to oGOLD, and belong to D groups (high risk, more symptoms) (45.6%), according to nGOLD. Overall, a low use of post-bronchodilator spirometry (65.1%) and of pulmonary rehabilitation (13.4%) was shown. The results highlighted a low prescriptive appropriateness but with higher value according to nGOLD than oGOLD: 61.4% vs 35.6%. Study limitations: Prescription data only provide limited information to judge prescribing quality, thus the results have to be evaluated with caution; moreover, this study was not designed to assess the difference between oGOLD and nGOLD. Conclusions: Guideline recommendations are applied only partially within clinical practice. A higher prescriptive appropriateness is shown by GPs using nGOLD classification. This might be due to the fact that nGOLD, with respect to oGOLD, takes into account anamnestic usual features considered by GPs in their clinical practice.
Introduction Chronic obstructive pulmonary disease (COPD) has been described by the World Health Organization Global Obstructive Lung Disease initiative and ! 2014 Informa UK Ltd www.cmrojournal.com
COPD management by general practitioners Maio et al.
1033
Curr Med Res Opin Downloaded from informahealthcare.com by Nyu Medical Center on 02/19/15 For personal use only.
Current Medical Research & Opinion Volume 30, Number 6
June 2014
by the American Thoracic Society (ATS)/European Respiratory Society (ERS) COPD guidelines as a disease ‘characterized by airflow limitation that is not fully reversible’1,2. COPD has emerged as the most important respiratory disease globally and its prevalence and impact on the society is expected to increase until it becomes the third ranking cause of death by 2030 worldwide3. The Global Strategy for the Diagnosis, Management and Prevention of Chronic Obstructive Pulmonary Disease (GOLD) report is one of the most frequently used documents for managing COPD patients worldwide1. Stratification of patients according to COPD severity is a crucial aspect in initiating treatment. Appropriate COPD management could reduce its impact on patients’ quality of life and health services costs1. General practitioners (GPs) are among the first healthcare professionals to whom patients refer for their symptoms and therefore GPs are encouraged to understand and use guidelines. Nevertheless, there are few studies concerning COPD management based on data directly collected in the GP’s office: they show that guidelines are only partially applied within clinical practice4,5 and often are not even read by GPs6. As a matter of fact, in 2005, an Italian survey on GPs showed that 30% of patients with a GP’s diagnosis of COPD had no spirometry; the main GPs’ reported reason for not performing spirometry was ‘because clinical assessment was enough’7. In the previous GOLD version, the severity classification was mainly based on the level of the airway obstruction (AO)8. In the 2011 GOLD update, important changes in the severity classification have been introduced (patient’s level of symptoms and history of exacerbations), determining a possible positive implication in the disease management1. The COMODHES study (Italian acronym for ‘Global assessment of COPD burden: mortality and morbidity in different health systems’), carried out in 2009–12, had the overall aim of assessing, at national level, the real impact of COPD on general population samples, combining information derived from current health databases and GPs. Data collected within the COMODHES study were analyzed aiming to compare COPD management by Italian GPs according to either the old GOLD (oGOLD) or the new GOLD (nGOLD) guidelines.
Methods Population Data regard GPs and their patients affected by COPD. GPs were distributed in the three Italian macro-areas: North, Center, South/Islands. GP recruitment was performed via mail among GPs that participated in previous studies or educational courses organized by the authors’ institution or 1034
COPD management by general practitioners Maio et al.
among GPs belonging to the Italian Society of General Medicine (SIMG). GPs were invited to select 15 patients with a COPD diagnosis at different disease severity levels. Patients provided ex-ante their written informed consent for the purposes of the study (participation in the study, anonymous management of individual data, collective statistical analyses and anonymous publication of research results). The study protocol, patient information sheet and consent form were approved by the Ethics Committee of the University-Hospital of Pisa on August 3, 2009 (Prot. no. 44538).
Data collection GPs filled in a questionnaire for each of the 15 selected patients to gather information on patients characteristics, COPD diagnosis, symptoms, disease severity, exacerbations, prescribed pharmacological and non-pharmacological treatments. In particular, GPs reported: the disease diagnosis based on their clinical experience and on clinical information about the patients, including spirometry; the symptoms level, according to the modified Medical Research Council (mMRC) scale9; the number of exacerbations in the last 12 months. COPD severity was computed on the basis of the information reported by GPs in the questionnaire: (a) according to oGOLD8, on the basis of the spirometric values; (b) according to nGOLD1, on the basis of spirometric values, symptoms level and number of exacerbations (Table 1). The appropriate COPD treatment was evaluated considering the pharmacological treatment recommended in the GOLD guidelines for the different severity levels (Table 2). The appropriateness to the regular treatment was computed independently from the use of short term bronchodilators, that are given to the patients on an as-needed basis for each severity level.
Statistical analyses Statistical analyses were carried out using the Statistical Package for the Social Sciences (SPSS), version 16.0. Routines used were frequency distribution and crosstabulations. The chi-square test was used to compare the GPs’ COPD management by oGOLD and nGOLD severity levels. The severe and very severe levels of the oGOLD were summed up due to the scarce number of subjects in the very severe level (n ¼ 18). A logistic regression model was applied to assess the association between lack of prescriptive appropriateness and oGOLD and nGOLD severity levels adjusting for the effects of gender, age, smoking habits, control of symptoms, number of exacerbations and health services use www.cmrojournal.com ! 2014 Informa UK Ltd
Current Medical Research & Opinion Volume 30, Number 6
June 2014
Table 1. oGOLD and nGOLD COPD severity classification. oGOLD8 Mild
Moderate
FEV1% pred 80
804 FEV1% pred 50
Curr Med Res Opin Downloaded from informahealthcare.com by Nyu Medical Center on 02/19/15 For personal use only.
nGOLD
Severe
Very severe
504 FEV1% pred 30
FEV1% pred 530
1
Group A
Group B
Group C
Group D
mild or moderate AO and/or 0–1 exacerbations for year and mMRC grade 0–1
mild or moderate AO and/or 0–1 exacerbations for year and mMRC grade 2
severe or very severe AO and/or 2 exacerbations for year and mMRC grade 0–1
severe or very severe AO and/or 2 exacerbations for year and mMRC grade 2
FEV1% pred: Forced Expiratory Volume in the first second in percentage of the predicted value; AO: airflow obstruction; mMRC: modified Medical Research Council scale for dyspnea.
Table 2. Summary of recommendations for COPD treatment according to GOLD guidelines. GOLD classification oGOLD classification8: mild moderate severe very severe nGOLD classification1: A B C D
Suggested pharmacological treatment
No therapy or SABA LABA or LAMA or theophylline LABA and/or LAMA þ ICS (with at least 1 exacerbation per year) LABA and/or LAMA þ ICS (with at least 1 exacerbation per year) SABA and/or SAMA or LABA or LAMA or theophylline LABA and/or LAMA or theophylline LAMA or LABA þ ICS or LABA þ LAMA or theophylline LAMA or (LABA and/or LAMA) þ ICS or LABA þ LAMA or theophylline
SABA: short-acting bronchodilator; LABA: long-acting beta-2 agonist; LAMA: long-acting muscarinic antagonist; ICS: inhaled corticosteroids.
(hospitalization, emergency room visits, specialists, GPs visits). Two-tailed tests at a 5% level of significance were applied in the analyses.
Results GP and patient characteristics GPs participating in the study (n ¼ 176) were mostly males (82.4%). Distribution among the macro-areas was: 42.6% in the Northern area, 23.9% in the Central area, and 33.5% in the South/Islands. On average, each GP took care of 1414 patients, of whom 40 suffered from COPD; 89.9% of GPs declared knowing national/ international guidelines for COPD management (Table 3). GPs filled in 2288 questionnaires but only 946 questionnaires reported spirometric values (41.3%): of ! 2014 Informa UK Ltd www.cmrojournal.com
Table 3. Descriptive characteristics of the GPs (valid percentage) (n ¼ 176). Variable Sex %: males females Italian geographic area %: North Center South/Islands Number of patients (mean SD) Number of COPD patients (mean SD) Knowledge of COPD guidelines %
82.4 17.6 42.6 23.9 33.5 1414 366 40 25 89.9
SD: standard deviation.
these 621 (65.6%) had AO confirmed from the spirometric values reported by GPs in the questionnaire. To compare the oGOLD with the nGOLD in the same subsample, analyses were performed in subjects having complete information on AO, mMRC scale and number of exacerbations (n ¼ 526). Comparing the symptoms level (mMRC scale) and the exacerbations number between included and excluded subjects, no statistical difference was found. Table 4 reports the general characteristics of patients. COPD patients had a mean age of 72.5 years and were mainly males (71.2%). Subjects were more frequently ex-smokers (44.4%) and with a low education level (elementary/no education: 77.6%). GPs reported sufficient disease control (based on their clinical experience or on pre-existing clinical/functional information) in 47.2% of the subjects with over two exacerbations in the last 12 months. More than 70% of patients received education on disease management or risk factors; 92.2% used health services due to COPD and almost every patient had COPD treatment (95.2%); 54.2% of patients had regular supervision from the specialist. COPD management by general practitioners Maio et al.
1035
Current Medical Research & Opinion Volume 30, Number 6
June 2014
Table 4. Descriptive characteristics of the patients (valid percentage) (n ¼ 526).
60 50
Variable
Curr Med Res Opin Downloaded from informahealthcare.com by Nyu Medical Center on 02/19/15 For personal use only.
%
Age (mean SD) (n ¼ 526) Age range Group of age %: 565 yrs (n ¼ 104) 65–74 yrs (n ¼ 178) 474 yrs (n ¼ 241) Gender %: Male (n ¼ 374) Female (n ¼ 151) Smoking habits %: Smokers (n ¼ 164) Ex-smokers (n ¼ 228) Non-smokers (n ¼ 122) Education %: No þ elementary (n ¼ 227) Middle high school (n ¼ 116) High school (n ¼ 70) College (n ¼ 29) Symptoms control %: Insufficient (n ¼ 97) Sufficient (n ¼ 245) Good (n ¼ 158) Optimum (n ¼ 19) Number of exacerbations last 12 months (mean SD) (n ¼ 526) Education to disease management % (n ¼ 526) Education to risk factors % (n ¼ 526) Use of health services* % (n ¼ 526) Pharmacological treatment for COPD % (n ¼ 526) Specialist supervising % (n ¼ 487)
72.5 9.8 25–94
30
19.9 34.0 46.1
10
71.2 28.8
51.4 26.2 15.8 6.6
Severe/very severe
0
Figure 1. COPD severity levels according to oGOLD.
50 40 Group A 30
Group B
%
Group C
20
18.7 47.2 30.4 3.7 2.3 2.0 87.4 71.9 92.2 95.2 54.2
Most patients (51.5%) had moderate COPD, according to oGOLD, and belonged to D group (45.6%), according to nGOLD (Figures 1 and 2).
Diagnosis and nonpharmacological treatment Spirometry had been recommended by GPs in more than 80% of the sample, but about a quarter were prebronchodilator spirometry. There were no differences according to the oGOLD and nGOLD (Table 5). The influenza vaccination was performed in 72.1% of patients and pneumococcal vaccination in 42.2%; pulmonary rehabilitation was offered to 13.4% of patients. Oxygen therapy was performed by 18.6% of all patients. In particular, 4.2% of mild patients and 12% of moderate patients (oGOLD) were prescribed oxygen therapy although not recommended by guidelines; in contrast, none of the group A patients and 14.3% of group B patients (nGOLD) were prescribed oxygen therapy (Table 5). Overall, the percentage of subjects undergoing these treatments increased with disease severity according to oGOLD and nGOLD. COPD management by general practitioners Maio et al.
Moderate
20
31.9 44.4 23.7
SD: standard deviation. *Use of at least one of the following: hospitalization, emergency room visit, specialist and GP visit due to COPD.
1036
Mild
40
Group D 10 0
Figure 2. COPD severity levels according to nGOLD.
Pharmacological treatment The more frequently prescribed drugs were only LABA and/or LAMA (20%), ICS þ LABA and/or LAMA (68.9%), in the whole sample. Considering oGOLD, over-treatment was shown in the mild/moderate level (31.4% consumption of LABA and/or LAMA in mild subjects, 44.8% of ICS þ LABA and/or LAMA in mild and 70.8% of ICS þ LABA and/or LAMA in moderate). Under-treatment was shown in severe/ very severe disease (11.5% of LABA and/or LAMA and 2.7% of no drugs or only vaccination) (Table 6). Considering nGOLD, over-treatment was shown in low risk groups (A–B) (5.2% consumption of only LABA þ LAMA in A, 45.6% of ICS þ LABA and/ or LAMA in A and 66.7% in B); 30.5% of C subjects were prescribed with LABA þLAMA þ ICS (only recommended for group D by the guidelines). An ineligible under-treatment was shown in group A (14.7% of no drugs or only vaccination) (Table 6). Figure 3 reports the assessment of prescriptive appropriateness, showing a higher value of appropriateness according to nGOLD (61.4% vs 35.6%) and a higher value of over-treatment according to oGOLD (52.8% vs 25.0%). GPs’ adherence to GOLD treatment in accordance with severity classification was higher in the more severe patients than in the mild ones: the lowest adherence was found for mild COPD (13.3%) and group B www.cmrojournal.com ! 2014 Informa UK Ltd
Current Medical Research & Opinion Volume 30, Number 6
June 2014
Table 5. Spirometry and non-pharmacological treatments by severity levels. oGOLD severity
N Only pre-bronchodilator (%) Post-bronchodilator (%) No spirometry (%) Influenza vaccination (%) Pneumococcal vaccination (%) Pulmonary rehabilitation (%) Oxygen therapy (%)
Mild
Moderate
Severe/very severe
Total
p-Value*
106 25.5 65.1 9.4 62.3 32.1 5.2 4.2
271 21.6 65.3 13.1 72.3 38.0 10.8 12.0
149 23.6 64.9 11.5 78.5 57.0 24.2 41.7
526 23.0 65.1 11.9 72.1 42.2 13.4 18.6
0.017 50.001 50.001 50.001
A
B
C
D
Total
p-Value*
117 26.7 57.8 15.5 54.3 25.9 2.8 0
51 17.6 68.6 13.7 66.7 33.3 6.1 14.3
118 26.3 64.9 8.8 70.4 37.4 9.8 4.9
240 20.4 68.5 11.1 82.4 54.2 21.3 35.2
526 23.0 65.1 11.9 72.1 42.2 13.4 18.6
50.001 50.001 50.001 50.001
Total
p-Value*
0.845
Curr Med Res Opin Downloaded from informahealthcare.com by Nyu Medical Center on 02/19/15 For personal use only.
nGOLD severity
N Only pre-bronchodilator (%) Post-bronchodilator (%) No spirometry (%) Influenza vaccination (%) Pneumococcal vaccination (%) Pulmonary rehabilitation (%) Oxygen therapy (%)
0.365
*Statistical difference between different severity levels by chi-square test.
Table 6. Pharmacological treatment for COPD by severity levels. oGOLD severity
N Only LABA (%) Only LAMA (%) Only LABA þ LAMA (%) ICS þ LABA or LAMA (%) ICS þ LABA þ LAMA (%) Only SBD (%) Other** (%) No drugs or only vaccination (%)
Mild
Moderate
106 11.4 18.1 1.9 26.7 18.1 2.9 10.5 10.5
271 5.2 6.6 8.5 31.7 39.1 1.5 3.7 3.7
Severe/very severe 149 5.4 2.7 3.4 24.5 57.8 0 3.4 2.7
526 6.5 7.8 5.7 28.7 40.2 1.3 5.0 4.8
50.001
nGOLD severity
N Only LABA (%) Only LAMA (%) Only LABA þ LAMA (%) ICS þ LABA or LAMA (%) ICS þ LABA þ LAMA (%) Only SBD (%) Other** (%) No drugs or only vaccination (%)
A
B
C
D
Total
117 10.3 16.4 5.2 28.4 17.2 2.6 5.2 14.7
51 3.9 5.9 7.8 35.3 31.4 3.9 3.9 7.8
118 5.9 11.0 7.6 33.9 30.5 0 8.5 2.5
240 5.5 2.5 4.6 24.8 58.0 0.8 3.4 0.4
526 6.5 7.8 5.7 28.7 40.2 1.3 5.0 4.8
p-Value* 50.001
LABA: long-acting beta-2 agonist; LAMA: long-acting muscarinic antagonist; ICS: inhaled corticosteroids; SBD: short-acting bronchodilator. *Statistical difference between different severity levels by chi-square test. **Other: other drugs not recommended in guidelines tables (only ICS, only antibiotic, only antioxidant, only mucolytics, mucolytics þ systemic corticosteroids, mucolytics þ antioxidant).
patients (21.6%), whilst the highest adherence was for severe/very severe COPD (79.6%) and for group D patients (90.8%). The results of the logistic regression analyses showed that, also adjusting for age, gender, smoking habits, ! 2014 Informa UK Ltd www.cmrojournal.com
number of exacerbations, control of symptoms and health services use, severe/very severe COPD, compared to mild COPD, and group D patients, compared to group A, were significant protective factors for lack of prescriptive appropriateness: odds ratio (OR) 0.09 (95% COPD management by general practitioners Maio et al.
1037
Current Medical Research & Opinion Volume 30, Number 6
June 2014
(a) 100
***
90 80 70 60 %
50
Appropriate
40
Over-treated
30
Under-treated
20
Not recommended drugs§
10
ta l
Se
ve
re /V e
M
ry
od
To
se ve re
er at e
ild M
Curr Med Res Opin Downloaded from informahealthcare.com by Nyu Medical Center on 02/19/15 For personal use only.
0
*** p
Vol. 30, No. 6, 2014, 1033–1042
Article ST-0372.R1/884492 All rights reserved: reproduction in whole or part not permitted
Curr Med Res Opin Downloaded from informahealthcare.com by Nyu Medical Center on 02/19/15 For personal use only.
Original article COPD management according to old and new GOLD guidelines: an observational study with Italian general practitioners
S. Maio S. Baldacci F. Martini S. Cerrai G. Sarno M. Borbotti Pulmonary Environmental Epidemiology Unit, CNR Institute of Clinical Physiology, Pisa, Italy
Abstract Background: Guideline recommendations for COPD management are only partially applied within primary care clinical practice. Objective: To compare the COPD management by Italian general practitioners (GPs) according to either the old GOLD (oGOLD) or the new GOLD (nGOLD) guidelines.
A.P. Pala Technosciences Unit, CNR Institute of Clinical Physiology, Pisa, Italy
N. Murgia Section of Occupational Medicine, Respiratory Diseases and Toxicology, University of Perugia, Perugia, Italy
G. Viegi Pulmonary Environmental Epidemiology Unit, CNR Institute of Clinical Physiology, Pisa, Italy
on behalf of the COMODHES study group Address for correspondence: Dr. Sara Maio, CNR Institute of Clinical Physiology, Via Trieste, 41 - 56126 Pisa, Italy. Tel: +39 050-502031; Fax: +39 050-503596; [email protected] Keywords: COPD – General practitioners – GOLD guidelines – Prescriptions pattern Accepted: 6 January 2014; published online: 3 February 2014 Citation: Curr Med Res Opin 2014; 30:1033–42
Research design and methods: Observational study in different Italian areas. A total of 176 GPs enrolled their patients with a COPD diagnosis. Questionnaires were used to collect data on: COPD symptoms, disease severity, exacerbations, prescribed pharmacological and non-pharmacological treatments. COPD severity was estimated according to oGOLD and nGOLD guidelines. Results: A total of 526 subjects had complete information to assess COPD severity level according to guidelines (symptoms level, spirometry, history of exacerbations). The investigated subjects were more frequently males (71.2%) with a mean age of 72.5 years, and ex-smokers (44.4%). GPs reported sufficient control of the disease in 47.2% of the subjects with over two exacerbations in the last 12 months. Most patients have moderate COPD (51.5%), according to oGOLD, and belong to D groups (high risk, more symptoms) (45.6%), according to nGOLD. Overall, a low use of post-bronchodilator spirometry (65.1%) and of pulmonary rehabilitation (13.4%) was shown. The results highlighted a low prescriptive appropriateness but with higher value according to nGOLD than oGOLD: 61.4% vs 35.6%. Study limitations: Prescription data only provide limited information to judge prescribing quality, thus the results have to be evaluated with caution; moreover, this study was not designed to assess the difference between oGOLD and nGOLD. Conclusions: Guideline recommendations are applied only partially within clinical practice. A higher prescriptive appropriateness is shown by GPs using nGOLD classification. This might be due to the fact that nGOLD, with respect to oGOLD, takes into account anamnestic usual features considered by GPs in their clinical practice.
Introduction Chronic obstructive pulmonary disease (COPD) has been described by the World Health Organization Global Obstructive Lung Disease initiative and ! 2014 Informa UK Ltd www.cmrojournal.com
COPD management by general practitioners Maio et al.
1033
Curr Med Res Opin Downloaded from informahealthcare.com by Nyu Medical Center on 02/19/15 For personal use only.
Current Medical Research & Opinion Volume 30, Number 6
June 2014
by the American Thoracic Society (ATS)/European Respiratory Society (ERS) COPD guidelines as a disease ‘characterized by airflow limitation that is not fully reversible’1,2. COPD has emerged as the most important respiratory disease globally and its prevalence and impact on the society is expected to increase until it becomes the third ranking cause of death by 2030 worldwide3. The Global Strategy for the Diagnosis, Management and Prevention of Chronic Obstructive Pulmonary Disease (GOLD) report is one of the most frequently used documents for managing COPD patients worldwide1. Stratification of patients according to COPD severity is a crucial aspect in initiating treatment. Appropriate COPD management could reduce its impact on patients’ quality of life and health services costs1. General practitioners (GPs) are among the first healthcare professionals to whom patients refer for their symptoms and therefore GPs are encouraged to understand and use guidelines. Nevertheless, there are few studies concerning COPD management based on data directly collected in the GP’s office: they show that guidelines are only partially applied within clinical practice4,5 and often are not even read by GPs6. As a matter of fact, in 2005, an Italian survey on GPs showed that 30% of patients with a GP’s diagnosis of COPD had no spirometry; the main GPs’ reported reason for not performing spirometry was ‘because clinical assessment was enough’7. In the previous GOLD version, the severity classification was mainly based on the level of the airway obstruction (AO)8. In the 2011 GOLD update, important changes in the severity classification have been introduced (patient’s level of symptoms and history of exacerbations), determining a possible positive implication in the disease management1. The COMODHES study (Italian acronym for ‘Global assessment of COPD burden: mortality and morbidity in different health systems’), carried out in 2009–12, had the overall aim of assessing, at national level, the real impact of COPD on general population samples, combining information derived from current health databases and GPs. Data collected within the COMODHES study were analyzed aiming to compare COPD management by Italian GPs according to either the old GOLD (oGOLD) or the new GOLD (nGOLD) guidelines.
Methods Population Data regard GPs and their patients affected by COPD. GPs were distributed in the three Italian macro-areas: North, Center, South/Islands. GP recruitment was performed via mail among GPs that participated in previous studies or educational courses organized by the authors’ institution or 1034
COPD management by general practitioners Maio et al.
among GPs belonging to the Italian Society of General Medicine (SIMG). GPs were invited to select 15 patients with a COPD diagnosis at different disease severity levels. Patients provided ex-ante their written informed consent for the purposes of the study (participation in the study, anonymous management of individual data, collective statistical analyses and anonymous publication of research results). The study protocol, patient information sheet and consent form were approved by the Ethics Committee of the University-Hospital of Pisa on August 3, 2009 (Prot. no. 44538).
Data collection GPs filled in a questionnaire for each of the 15 selected patients to gather information on patients characteristics, COPD diagnosis, symptoms, disease severity, exacerbations, prescribed pharmacological and non-pharmacological treatments. In particular, GPs reported: the disease diagnosis based on their clinical experience and on clinical information about the patients, including spirometry; the symptoms level, according to the modified Medical Research Council (mMRC) scale9; the number of exacerbations in the last 12 months. COPD severity was computed on the basis of the information reported by GPs in the questionnaire: (a) according to oGOLD8, on the basis of the spirometric values; (b) according to nGOLD1, on the basis of spirometric values, symptoms level and number of exacerbations (Table 1). The appropriate COPD treatment was evaluated considering the pharmacological treatment recommended in the GOLD guidelines for the different severity levels (Table 2). The appropriateness to the regular treatment was computed independently from the use of short term bronchodilators, that are given to the patients on an as-needed basis for each severity level.
Statistical analyses Statistical analyses were carried out using the Statistical Package for the Social Sciences (SPSS), version 16.0. Routines used were frequency distribution and crosstabulations. The chi-square test was used to compare the GPs’ COPD management by oGOLD and nGOLD severity levels. The severe and very severe levels of the oGOLD were summed up due to the scarce number of subjects in the very severe level (n ¼ 18). A logistic regression model was applied to assess the association between lack of prescriptive appropriateness and oGOLD and nGOLD severity levels adjusting for the effects of gender, age, smoking habits, control of symptoms, number of exacerbations and health services use www.cmrojournal.com ! 2014 Informa UK Ltd
Current Medical Research & Opinion Volume 30, Number 6
June 2014
Table 1. oGOLD and nGOLD COPD severity classification. oGOLD8 Mild
Moderate
FEV1% pred 80
804 FEV1% pred 50
Curr Med Res Opin Downloaded from informahealthcare.com by Nyu Medical Center on 02/19/15 For personal use only.
nGOLD
Severe
Very severe
504 FEV1% pred 30
FEV1% pred 530
1
Group A
Group B
Group C
Group D
mild or moderate AO and/or 0–1 exacerbations for year and mMRC grade 0–1
mild or moderate AO and/or 0–1 exacerbations for year and mMRC grade 2
severe or very severe AO and/or 2 exacerbations for year and mMRC grade 0–1
severe or very severe AO and/or 2 exacerbations for year and mMRC grade 2
FEV1% pred: Forced Expiratory Volume in the first second in percentage of the predicted value; AO: airflow obstruction; mMRC: modified Medical Research Council scale for dyspnea.
Table 2. Summary of recommendations for COPD treatment according to GOLD guidelines. GOLD classification oGOLD classification8: mild moderate severe very severe nGOLD classification1: A B C D
Suggested pharmacological treatment
No therapy or SABA LABA or LAMA or theophylline LABA and/or LAMA þ ICS (with at least 1 exacerbation per year) LABA and/or LAMA þ ICS (with at least 1 exacerbation per year) SABA and/or SAMA or LABA or LAMA or theophylline LABA and/or LAMA or theophylline LAMA or LABA þ ICS or LABA þ LAMA or theophylline LAMA or (LABA and/or LAMA) þ ICS or LABA þ LAMA or theophylline
SABA: short-acting bronchodilator; LABA: long-acting beta-2 agonist; LAMA: long-acting muscarinic antagonist; ICS: inhaled corticosteroids.
(hospitalization, emergency room visits, specialists, GPs visits). Two-tailed tests at a 5% level of significance were applied in the analyses.
Results GP and patient characteristics GPs participating in the study (n ¼ 176) were mostly males (82.4%). Distribution among the macro-areas was: 42.6% in the Northern area, 23.9% in the Central area, and 33.5% in the South/Islands. On average, each GP took care of 1414 patients, of whom 40 suffered from COPD; 89.9% of GPs declared knowing national/ international guidelines for COPD management (Table 3). GPs filled in 2288 questionnaires but only 946 questionnaires reported spirometric values (41.3%): of ! 2014 Informa UK Ltd www.cmrojournal.com
Table 3. Descriptive characteristics of the GPs (valid percentage) (n ¼ 176). Variable Sex %: males females Italian geographic area %: North Center South/Islands Number of patients (mean SD) Number of COPD patients (mean SD) Knowledge of COPD guidelines %
82.4 17.6 42.6 23.9 33.5 1414 366 40 25 89.9
SD: standard deviation.
these 621 (65.6%) had AO confirmed from the spirometric values reported by GPs in the questionnaire. To compare the oGOLD with the nGOLD in the same subsample, analyses were performed in subjects having complete information on AO, mMRC scale and number of exacerbations (n ¼ 526). Comparing the symptoms level (mMRC scale) and the exacerbations number between included and excluded subjects, no statistical difference was found. Table 4 reports the general characteristics of patients. COPD patients had a mean age of 72.5 years and were mainly males (71.2%). Subjects were more frequently ex-smokers (44.4%) and with a low education level (elementary/no education: 77.6%). GPs reported sufficient disease control (based on their clinical experience or on pre-existing clinical/functional information) in 47.2% of the subjects with over two exacerbations in the last 12 months. More than 70% of patients received education on disease management or risk factors; 92.2% used health services due to COPD and almost every patient had COPD treatment (95.2%); 54.2% of patients had regular supervision from the specialist. COPD management by general practitioners Maio et al.
1035
Current Medical Research & Opinion Volume 30, Number 6
June 2014
Table 4. Descriptive characteristics of the patients (valid percentage) (n ¼ 526).
60 50
Variable
Curr Med Res Opin Downloaded from informahealthcare.com by Nyu Medical Center on 02/19/15 For personal use only.
%
Age (mean SD) (n ¼ 526) Age range Group of age %: 565 yrs (n ¼ 104) 65–74 yrs (n ¼ 178) 474 yrs (n ¼ 241) Gender %: Male (n ¼ 374) Female (n ¼ 151) Smoking habits %: Smokers (n ¼ 164) Ex-smokers (n ¼ 228) Non-smokers (n ¼ 122) Education %: No þ elementary (n ¼ 227) Middle high school (n ¼ 116) High school (n ¼ 70) College (n ¼ 29) Symptoms control %: Insufficient (n ¼ 97) Sufficient (n ¼ 245) Good (n ¼ 158) Optimum (n ¼ 19) Number of exacerbations last 12 months (mean SD) (n ¼ 526) Education to disease management % (n ¼ 526) Education to risk factors % (n ¼ 526) Use of health services* % (n ¼ 526) Pharmacological treatment for COPD % (n ¼ 526) Specialist supervising % (n ¼ 487)
72.5 9.8 25–94
30
19.9 34.0 46.1
10
71.2 28.8
51.4 26.2 15.8 6.6
Severe/very severe
0
Figure 1. COPD severity levels according to oGOLD.
50 40 Group A 30
Group B
%
Group C
20
18.7 47.2 30.4 3.7 2.3 2.0 87.4 71.9 92.2 95.2 54.2
Most patients (51.5%) had moderate COPD, according to oGOLD, and belonged to D group (45.6%), according to nGOLD (Figures 1 and 2).
Diagnosis and nonpharmacological treatment Spirometry had been recommended by GPs in more than 80% of the sample, but about a quarter were prebronchodilator spirometry. There were no differences according to the oGOLD and nGOLD (Table 5). The influenza vaccination was performed in 72.1% of patients and pneumococcal vaccination in 42.2%; pulmonary rehabilitation was offered to 13.4% of patients. Oxygen therapy was performed by 18.6% of all patients. In particular, 4.2% of mild patients and 12% of moderate patients (oGOLD) were prescribed oxygen therapy although not recommended by guidelines; in contrast, none of the group A patients and 14.3% of group B patients (nGOLD) were prescribed oxygen therapy (Table 5). Overall, the percentage of subjects undergoing these treatments increased with disease severity according to oGOLD and nGOLD. COPD management by general practitioners Maio et al.
Moderate
20
31.9 44.4 23.7
SD: standard deviation. *Use of at least one of the following: hospitalization, emergency room visit, specialist and GP visit due to COPD.
1036
Mild
40
Group D 10 0
Figure 2. COPD severity levels according to nGOLD.
Pharmacological treatment The more frequently prescribed drugs were only LABA and/or LAMA (20%), ICS þ LABA and/or LAMA (68.9%), in the whole sample. Considering oGOLD, over-treatment was shown in the mild/moderate level (31.4% consumption of LABA and/or LAMA in mild subjects, 44.8% of ICS þ LABA and/or LAMA in mild and 70.8% of ICS þ LABA and/or LAMA in moderate). Under-treatment was shown in severe/ very severe disease (11.5% of LABA and/or LAMA and 2.7% of no drugs or only vaccination) (Table 6). Considering nGOLD, over-treatment was shown in low risk groups (A–B) (5.2% consumption of only LABA þ LAMA in A, 45.6% of ICS þ LABA and/ or LAMA in A and 66.7% in B); 30.5% of C subjects were prescribed with LABA þLAMA þ ICS (only recommended for group D by the guidelines). An ineligible under-treatment was shown in group A (14.7% of no drugs or only vaccination) (Table 6). Figure 3 reports the assessment of prescriptive appropriateness, showing a higher value of appropriateness according to nGOLD (61.4% vs 35.6%) and a higher value of over-treatment according to oGOLD (52.8% vs 25.0%). GPs’ adherence to GOLD treatment in accordance with severity classification was higher in the more severe patients than in the mild ones: the lowest adherence was found for mild COPD (13.3%) and group B www.cmrojournal.com ! 2014 Informa UK Ltd
Current Medical Research & Opinion Volume 30, Number 6
June 2014
Table 5. Spirometry and non-pharmacological treatments by severity levels. oGOLD severity
N Only pre-bronchodilator (%) Post-bronchodilator (%) No spirometry (%) Influenza vaccination (%) Pneumococcal vaccination (%) Pulmonary rehabilitation (%) Oxygen therapy (%)
Mild
Moderate
Severe/very severe
Total
p-Value*
106 25.5 65.1 9.4 62.3 32.1 5.2 4.2
271 21.6 65.3 13.1 72.3 38.0 10.8 12.0
149 23.6 64.9 11.5 78.5 57.0 24.2 41.7
526 23.0 65.1 11.9 72.1 42.2 13.4 18.6
0.017 50.001 50.001 50.001
A
B
C
D
Total
p-Value*
117 26.7 57.8 15.5 54.3 25.9 2.8 0
51 17.6 68.6 13.7 66.7 33.3 6.1 14.3
118 26.3 64.9 8.8 70.4 37.4 9.8 4.9
240 20.4 68.5 11.1 82.4 54.2 21.3 35.2
526 23.0 65.1 11.9 72.1 42.2 13.4 18.6
50.001 50.001 50.001 50.001
Total
p-Value*
0.845
Curr Med Res Opin Downloaded from informahealthcare.com by Nyu Medical Center on 02/19/15 For personal use only.
nGOLD severity
N Only pre-bronchodilator (%) Post-bronchodilator (%) No spirometry (%) Influenza vaccination (%) Pneumococcal vaccination (%) Pulmonary rehabilitation (%) Oxygen therapy (%)
0.365
*Statistical difference between different severity levels by chi-square test.
Table 6. Pharmacological treatment for COPD by severity levels. oGOLD severity
N Only LABA (%) Only LAMA (%) Only LABA þ LAMA (%) ICS þ LABA or LAMA (%) ICS þ LABA þ LAMA (%) Only SBD (%) Other** (%) No drugs or only vaccination (%)
Mild
Moderate
106 11.4 18.1 1.9 26.7 18.1 2.9 10.5 10.5
271 5.2 6.6 8.5 31.7 39.1 1.5 3.7 3.7
Severe/very severe 149 5.4 2.7 3.4 24.5 57.8 0 3.4 2.7
526 6.5 7.8 5.7 28.7 40.2 1.3 5.0 4.8
50.001
nGOLD severity
N Only LABA (%) Only LAMA (%) Only LABA þ LAMA (%) ICS þ LABA or LAMA (%) ICS þ LABA þ LAMA (%) Only SBD (%) Other** (%) No drugs or only vaccination (%)
A
B
C
D
Total
117 10.3 16.4 5.2 28.4 17.2 2.6 5.2 14.7
51 3.9 5.9 7.8 35.3 31.4 3.9 3.9 7.8
118 5.9 11.0 7.6 33.9 30.5 0 8.5 2.5
240 5.5 2.5 4.6 24.8 58.0 0.8 3.4 0.4
526 6.5 7.8 5.7 28.7 40.2 1.3 5.0 4.8
p-Value* 50.001
LABA: long-acting beta-2 agonist; LAMA: long-acting muscarinic antagonist; ICS: inhaled corticosteroids; SBD: short-acting bronchodilator. *Statistical difference between different severity levels by chi-square test. **Other: other drugs not recommended in guidelines tables (only ICS, only antibiotic, only antioxidant, only mucolytics, mucolytics þ systemic corticosteroids, mucolytics þ antioxidant).
patients (21.6%), whilst the highest adherence was for severe/very severe COPD (79.6%) and for group D patients (90.8%). The results of the logistic regression analyses showed that, also adjusting for age, gender, smoking habits, ! 2014 Informa UK Ltd www.cmrojournal.com
number of exacerbations, control of symptoms and health services use, severe/very severe COPD, compared to mild COPD, and group D patients, compared to group A, were significant protective factors for lack of prescriptive appropriateness: odds ratio (OR) 0.09 (95% COPD management by general practitioners Maio et al.
1037
Current Medical Research & Opinion Volume 30, Number 6
June 2014
(a) 100
***
90 80 70 60 %
50
Appropriate
40
Over-treated
30
Under-treated
20
Not recommended drugs§
10
ta l
Se
ve
re /V e
M
ry
od
To
se ve re
er at e
ild M
Curr Med Res Opin Downloaded from informahealthcare.com by Nyu Medical Center on 02/19/15 For personal use only.
0
*** p
