A Real-World Evaluation of Therapeutic Inertia in Patients with Severe Uncontrolled Chronic Obstructive Pulmonary Disease Authors: Myriam Calle Rubio, Juan José Soler-Cataluña, Pere Almagro, Arturo Huerta, Diego González-Segura, Borja G Cosío Published: 2026 ABSTRACT BackgroundTherapeutic inertia (TI) defined as failure to escalate or initiate adequate therapy when treatment goals are not met, contributes to poor outcome in patients with chronic obstructive pulmonary disease (COPD). This real-world study aimed to investigate TI and the discrepancy between disease control assessed by physicians’ perception and objective methods.MethodsThis was a post hoc analysis of data of 4801 patients with severe COPD (FEV1 < 50% predicted) included in an observational cross-sectional multicenter nationwide study conducted in Spain. Controlled vs. uncontrolled disease was considered according to COPD assessment test (CAT) score ≤ or > 16 and absence or presence of at least one exacerbation in the past 3 months. A 5-point Likert scale (very poor/poor; satisfactory/good/very good) was used to assess COPD control perceived by physicians and patients’ satisfaction with treatment.ResultsUncontrolled COPD was present in 3479 of 4801 patients (72.5%) and controlled COPD in 1322 (27.5%). Among patients with uncontrolled COPD, no treatment adjustments were made in 1519, with a rate of TI of 43.7%. In the multivariate analysis, former smokers, lower exacerbations history, white sputum, lower degree of dyspnea, higher adherence to inhaled therapy, use of dual/triple inhaled medications, absence of SABA use, secondary healthcare level, and patient’s satisfaction with treatment were significantly associated with TI. Physicians perceived controlled disease in 2373 patients with uncontrolled COPD, with a discordance rate of 68.2%. Among patients with controlled COPD, the discordant rate was 6.3%. The most influential variables associated with controlled COPD rated by physicians were patients’ satisfaction with treatment, no exacerbations in the past year, triple therapy, and ex-smoking status.ConclusionThe high rate of TI and the gap between actual COPD control and physicians’ perceptions, highlights the urgent need to systematize and standardize the follow-up of patients with severe COPD in daily practice. Introduction Chronic obstructive pulmonary disease (COPD) continues to be a leading cause of morbimortality and healthcare burden due to the interplay of different factors, particularly, disease heterogeneity, complexity of symptoms, frequent exacerbations, and suboptimal treatment.1–3 Evidence-based management of patients with COPD according to recommendations of clinical practice guidelines is crucial for decreasing the impact of the disease and improving overall outcomes.4–6 However, even with the dissemination of guidelines, COPD remains mismanaged and underdiagnosed. Numerous barriers to COPD guideline usage have been identified, including poor familiarity with recommendations, mislabeling of COPD phenotypes, low self-efficacy, inability to incorporate spirometry into patient flow, improper selection of pharmacotherapy options, and time constraints.7–9 Moreover, real-work studies have shown a substantial variability in the adherence to guidelines depending on specific components of recommendations, patient population, COPD stages, and healthcare setting.10–12 Therapeutic inertia (TI), defined as failure to change therapy when treatment goals are not met, either by switching medication or increasing the dose, has been recognized as an important factor contributing to the gap between guidelines and the adequate management of COPD patients.13,14 In fact, TI is one facet of the broad concept of clinical inertia encompassing physician, patient, and health system factors, including a failure to assess risk, screen for and manage risk factors and complications, address nonpharmacological factors, and refer appropriately. Overcoming TI plays a key role in reducing mortality and morbidity in people with COPD, as well as in avoiding further exacerbations and worsening lung function.15 Although TI has been extensively studied in other chronic diseases, such as diabetes16 and hypertension,17 which has been shown to potentially account for 80% of cardiovascular events, TI related to the management of COPD has been less frequently analyzed. In a recent cross-sectional clinical audit (the EPOCONSUL audit) that evaluated the outpatient care provided to 4225 patients with a diagnosis of COPD in 45 respiratory clinics in Spain, there was TI in 68.4% of patients with uncontrolled COPD and no action was taken during the visit for 9.1% of cases.18 Moreover, TI was 3.3 times more likely when the attending physician reported the disease as controlled and in patients with less history of exacerbations.18 In order to provide further evidence of the effect of TI on the management COPD patients, a post hoc analysis was performed using data of a large population of patients with severe COPD included in a recent observational, cross-sectional, and multicenter study.19 Based on the hypothesis that TI may affect the control of COPD, we investigated the rate of TI in patients with severe uncontrolled COPD in real-world clinical practice. Materials and Methods Study Design and Patients The present study was a post hoc analysis of data recorded in The CLAVE Study (CLAVE is the Spanish acronym for Caracterizar La EPOC grAVE en España, To Characterize Severe COPD in Spain), which was an observational, cross-sectional and multicenter study involving 4801 patients with severe COPD. These patients were recruited during routine clinical visits in 406 centers throughout Spain between September 2017 and May 2018. Details of the study have been previously reported.19 Eligibility criteria included patients of both sexes, aged ≥ 40 years, active smokers or ex-smokers, diagnosed with COPD with post-bronchodilator FEV1 < 50% of predicted (at least 12 months before recruitment), and receiving pharmacological therapies. Patients treated with oral corticosteroids or antibiotics due to a COPD exacerbation at the time of enrollment were excluded. The primary efficacy outcome of the CLAVE study was to assess control of severe COPD using a validated clinical control criteria tool.19 The study complied with the Declaration of Helsinki. Institutional review committee approval was obtained from Hospital Clinic, Barcelona (Spain) on 26th of September 2017 and all patients provided written informed consent. Study Endpoints The primary endpoint was the rate of TI in patients with severe uncontrolled COPD. Secondary endpoints included the relationship between sociodemographic, clinical, and treatment-related factors and TI, and how physician’s perceived COPD control in patients with severe uncontrolled COPD. Definitions and Data Collection TI was defined as the lack of any treatment adjustment, including switching/adding medication or increasing doses, during the clinical visit in patients with severe uncontrolled COPD. Control of COPD was evaluated with the Spanish version of the COPD assessment test (CAT)20 combined with exacerbations, using a CAT cutoff score of ≤ or > 16 and considering the absence or presence of exacerbations in the past 3 months. Therefore, uncontrolled COPD was defined as a CAT score > 16 and at least one exacerbation episode within the past 3 months, whereas controlled COPD was defined as a CAT score ≤ 16 and no exacerbations in the past 3 months. For each patient, the following data were recorded: demographics (sex and age); body mass index (BMI); smoking status (categorized as current smoker, ex-smoker/never smoker); time since COPD diagnosis; working status (active, inactive including retired, sick leave, housewife); place of residence (rural < 2000 inhabitants, semiurban 2000–10,000 inhabitants, urban > 10,000 inhabitants); family/social support (yes, no); history of allergy/atopy, nasal polyposis, allergic rhinitis; level of care (primary, secondary); level of physical activity (categorized as mild, moderate, and low using the Spanish version of the International Physical Activity Questionnaire [IPAQ]21); Charlson comorbidity index;22 presence of dyspnea using the modified Medical Research Council (mMRC) scale;23 number of exacerbations in the previous year (severe and moderate); severity of airflow limitation (post-bronchodilation FEV1% predicted); eosinophil blood count (categorized as < 100, 100–300, > 300 cells/µL); spontaneous sputum color (white, yellow/green or other); adherence to inhaled therapy according to the test of the adherence to inhalers (TAI) (categorized as good, moderate, and poor);24 and pharmacological and non-pharmacological treatments. The physician’s perceived degree of COPD control was evaluated using a 5-point Likert scale, ranging from −2: very poor; −1: poor; 0: satisfactory; 1: good; and 2: very good. Good control was considered for scores 0 (satisfactory), 1 (good), and 2 (very good), and poor control for scores −2 (very poor) and −1 (poor). The same criteria were applied for patients’ satisfaction with treatment, with results categorized as “yes” (satisfied) (scored 0, 1, and 2) and “no” (dissatisfied) (scores −2 and −1). Statistical Analysis Categorical variables are expressed as frequencies and percentages, and quantitative variables as mean and standard deviation (SD). The chi-square test or the Fisher’s exact test was used for the comparison of categorical variables and the Mann–Whitney U-test for the comparison of continuous variables. All variables with a p value < 0.10 in the bivariate analysis were included in a binary logistic regression model to identify independent variables associated with TI and the discordance between the degree of COPD control perceived by physicians in patients with uncontrolled COPD. A backward stepwise selection procedure (stay criterion p < 0.05) was used. The odds ratio (OR) and 95% confidence intervals (CI) were calculated. No imputations for missing data were made. Adjustments for center variability were not performed. Statistical significance was set at p < 0.05. The SAS (Statistical Analysis Systems (SAS Institute, Cary, NC, USA) version 9.4 was used for statistical analysis. Results Therapeutic Inertia and Associated Variables Of a total of 4801 patients with severe COPD included in the study, uncontrolled COPD was present in 72.5% (n = 3479) patients and controlled COPD in the remaining 27.5% (n = 1322). Among patients with uncontrolled COPD, no treatment adjustments were made in 1519, with a rate of TI of 43.7%. The distribution of study variables in the presence or absence of TI is shown in Table 1. In patients with TI compared to those without TI, the percentages of males, ex-smokers/never smokers, high level of physical activity, inactive working status, white sputum color, and good/intermediate adherence to inhaled therapy were significantly higher, whereas the mean Charlson comorbidity index, mean mMRC score, patients with allergic rhinitis, and exacerbations in the last year were significantly lower. The patients’ perceived satisfaction with treatment was rated as satisfactory/good/very good in 93.5% of cases in the TI group vs. 76.7% in the group without TI (p < 0.0001). Details of pharmacological treatment among patients with and without TI are shown in Table 2. In relation to inhaled therapy, patients with TI compared to those without TI had significantly lower rates of monotherapy and short-acting beta-2 agonists (SABA) use, whereas dual and triple inhalation combinations were significantly more common.Table 1Variables Associated with Therapeutic Inertia (TI)VariablesPatients with Uncontrolled COPD (n = 3479)Absence of TI (n = 1960)Presence of TI (n = 1519)P valueSex, n (%) Males1575 (80.4)1269 (83.5)0.015 Females385 (19.6)250 (16.5)Age, years, mean (SD)69.8 (9.6)70.2 (9.1)0.265Body mass index, kg/m2, mean (SD)27.4 (5.3)27.5 (5.1)0.464Smoking, n (%) Smoker574 (29.3)261 (17.2)< 0.001 Ex-smoker/never smoker1386 (70.7)1258 (82.8)Years since COPD diagnosis, mean (SD)10.2 (7.1)9.8 (6.6)0.268Level of physical activity, n (%) High197 (10.1)225 (14.8)< 0.001 Moderate712 (36.3)599 (39.4) Low/inactive921 (47.0)615 (40.5) Missing130 (6.6)80 (5.3)Level of care, n (%) Primary160 (8.2)120 (7.9)0.041 Secondary (pulmonology, internal medicine)1766 (90.1)1369 (90.1) Missing34 (1.7)30 (2.0)Working status, n (%) Active297 (15.2)173 (11.4)0.001 Inactive1663 (84.8)1346 (88.6)Place of residence, n (%) Rural317 (16.2)267 (17.6)0.158 Semiurban367 (18.7)311 (20.5) Urban1276 (65.1)941 (61.9)Family/social support, n (%) Yes693 (35.4)494 (32.5)0.080 No1267 (64.6)1025 (67.5)Comorbidities, n (%) Allergy/atopy143 (7.3)106 (7.0)0.718 Nasal polyposis37 (1.9)18 (1.2)0.098 Allergic rhinitis150 (7.7)82 (5.4)0.008Charlson comorbidity index, mean (SD)2.4 (1.7)2.2 (1.5)0.003Dyspnea mMRC scale, mean (SD)2.4 (0.9)2.2 (1.0)< 0.001Exacerbations in the last year, n (%) Severe (≥1)929 (47.4)578 (38.1)< 0.001 Moderate (≥2)723 (36.9)423 (27.8)< 0.001Post-bronchodilator FEV1% predicted, mean (SD)38.3 (8.5)38.6 (8.7)0.147Sputum color, n (%) White1321 (67.4)1217 (80.1)< 0.001 Yellow/green638 (32.5)301 (19.8)Eosinophil blood count, cells/µL, n (%) <100355 (18.1)248 (16.3)0.048 100-300844 (43.1)742 (48.8) >300291 (14.8)191 (12.6) Missing470 (24.0)338 (22.2)TAI adherence to inhaled therapy, n (%) Good993 (50.7)924 (60.8)< 0.001 Intermediate345 (17.6)300 (19.7) Poor505 (25.8)221 (14.5) Missing117 (6.0)74 (4.9)Abbreviations: SD, standard deviation; mMRC, modified Medical Research Council; TAI, test of adherence to inhalers. Table 2Pharmacological Treatment and Therapeutic Inertia (TI)VariablesPatients with Uncontrolled COPD (n = 3479)Absence of TI (n = 1960)Presence of TI (n = 1519)P valueInhaled therapy, n (%)1943 (99.1)1515 (99.7) Monotherapy145 (7.5)44 (2.9)< 0.001 Long-acting beta2-agonist (LABA)24 (1.2)11 (0.7) Long-acting muscarinic antagonist (LAMA)94 (4.8)21 (1.4) Inhaled corticosteroids (ICS)27 (1.4)12 (0.8) Dual therapy778 (40.0)631 (41.5) LABA/LAMA531 (27.3)491 (32.4) LABA/ICS241 (12.4)133 (8.8) LAMA/ICS6 (0.3)7 (0.5) Triple therapy986 (50.7)838 (55.3) Missing34 (1.7)2 (0.1)Short-acting beta2-agonist (SABA), n (%) Yes506 (25.8)299 (19.7)< 0.001 No1454 (74.2)1220 (80.3)Home oxygen therapy, n (%) Yes643 (32.8)508 (33.4)0.699 No1316 (67.2)1011 (66.6) Table 1Variables Associated with Therapeutic Inertia (TI)VariablesPatients with Uncontrolled COPD (n = 3479)Absence of TI (n = 1960)Presence of TI (n = 1519)P valueSex, n (%) Males1575 (80.4)1269 (83.5)0.015 Females385 (19.6)250 (16.5)Age, years, mean (SD)69.8 (9.6)70.2 (9.1)0.265Body mass index, kg/m2, mean (SD)27.4 (5.3)27.5 (5.1)0.464Smoking, n (%) Smoker574 (29.3)261 (17.2)< 0.001 Ex-smoker/never smoker1386 (70.7)1258 (82.8)Years since COPD diagnosis, mean (SD)10.2 (7.1)9.8 (6.6)0.268Level of physical activity, n (%) High197 (10.1)225 (14.8)< 0.001 Moderate712 (36.3)599 (39.4) Low/inactive921 (47.0)615 (40.5) Missing130 (6.6)80 (5.3)Level of care, n (%) Primary160 (8.2)120 (7.9)0.041 Secondary (pulmonology, internal medicine)1766 (90.1)1369 (90.1) Missing34 (1.7)30 (2.0)Working status, n (%) Active297 (15.2)173 (11.4)0.001 Inactive1663 (84.8)1346 (88.6)Place of residence, n (%) Rural317 (16.2)267 (17.6)0.158 Semiurban367 (18.7)311 (20.5) Urban1276 (65.1)941 (61.9)Family/social support, n (%) Yes693 (35.4)494 (32.5)0.080 No1267 (64.6)1025 (67.5)Comorbidities, n (%) Allergy/atopy143 (7.3)106 (7.0)0.718 Nasal polyposis37 (1.9)18 (1.2)0.098 Allergic rhinitis150 (7.7)82 (5.4)0.008Charlson comorbidity index, mean (SD)2.4 (1.7)2.2 (1.5)0.003Dyspnea mMRC scale, mean (SD)2.4 (0.9)2.2 (1.0)< 0.001Exacerbations in the last year, n (%) Severe (≥1)929 (47.4)578 (38.1)< 0.001 Moderate (≥2)723 (36.9)423 (27.8)< 0.001Post-bronchodilator FEV1% predicted, mean (SD)38.3 (8.5)38.6 (8.7)0.147Sputum color, n (%) White1321 (67.4)1217 (80.1)< 0.001 Yellow/green638 (32.5)301 (19.8)Eosinophil blood count, cells/µL, n (%) <100355 (18.1)248 (16.3)0.048 100-300844 (43.1)742 (48.8) >300291 (14.8)191 (12.6) Missing470 (24.0)338 (22.2)TAI adherence to inhaled therapy, n (%) Good993 (50.7)924 (60.8)< 0.001 Intermediate345 (17.6)300 (19.7) Poor505 (25.8)221 (14.5) Missing117 (6.0)74 (4.9)Abbreviations: SD, standard deviation; mMRC, modified Medical Research Council; TAI, test of adherence to inhalers. Variables Associated with Therapeutic Inertia (TI) Abbreviations: SD, standard deviation; mMRC, modified Medical Research Council; TAI, test of adherence to inhalers. Table 2Pharmacological Treatment and Therapeutic Inertia (TI)VariablesPatients with Uncontrolled COPD (n = 3479)Absence of TI (n = 1960)Presence of TI (n = 1519)P valueInhaled therapy, n (%)1943 (99.1)1515 (99.7) Monotherapy145 (7.5)44 (2.9)< 0.001 Long-acting beta2-agonist (LABA)24 (1.2)11 (0.7) Long-acting muscarinic antagonist (LAMA)94 (4.8)21 (1.4) Inhaled corticosteroids (ICS)27 (1.4)12 (0.8) Dual therapy778 (40.0)631 (41.5) LABA/LAMA531 (27.3)491 (32.4) LABA/ICS241 (12.4)133 (8.8) LAMA/ICS6 (0.3)7 (0.5) Triple therapy986 (50.7)838 (55.3) Missing34 (1.7)2 (0.1)Short-acting beta2-agonist (SABA), n (%) Yes506 (25.8)299 (19.7)< 0.001 No1454 (74.2)1220 (80.3)Home oxygen therapy, n (%) Yes643 (32.8)508 (33.4)0.699 No1316 (67.2)1011 (66.6) Pharmacological Treatment and Therapeutic Inertia (TI) In the logistic regression analysis (Figure 1), variables independently associated with TI were former smokers, lower exacerbations history, white sputum, lower degree of dyspnea, higher adherence to inhaled therapy, use of dual or triple inhaled medications, absence of SABA use, and healthcare level. Also, the patients’ perceived satisfaction with treatment was the most influential factor for TI (OR 3.14, 95% CI 2.44–4.03) followed by triple inhaled therapy (OR 2.14, 95% CI 1.44–3.18). Figure 1Variables independently associated with therapeutic inertia in patients with severe uncontrolled COPD (CAT score > 16 and at least one exacerbation in the past 3 months); *p < 0.01; **p < 0.0001.The graph presents odds ratios with 95% confidence intervals for factors affecting therapeutic inertia in COPD patients. The x-axis, labeled 'Odds ratio (95% CI)', ranges from 0 to 4.5. Factors on the y-axis include: Internal Medicine vs. pulmonology (1.33, 1.06-1.68), Primary care vs. pulmonology (1.46, 1.08-1.97), SABA treatment: no vs. yes (1.56, 1.30-1.88), Triple inhaled therapy vs. monotherapy (2.14, 1.44-3.18), Dual inhaled therapy vs. monotherapy (1.73, 1.17-2.57), Patient satisfaction: yes vs. no (3.14, 2.44-4.03), Adherence to inhaled therapy: good vs. poor (1.87, 1.50-2.32), intermediate vs. poor (1.77, 1.38-2.27), Degree of dyspnea (mMRC) (0.85, 0.79-0.93), Sputum color: white vs. yellow/green (1.59, 1.32-1.91), Moderate exacerbation last year: <2 vs. ≥2 (1.34, 1.14-1.58), Severe exacerbations last year: 0 vs. ≥1 (1.38, 1.18-1.61), Smoking status: ex/never smoker vs. smoker (1.89, 1.57-2.28). Asterisks denote statistical significance.A graph showing odds ratios for various factors affecting therapeutic inertia in severe COPD patients.Abbreviations: CI, confidence interval; SABA, short-acting beta2-agonist; mMRC, modified Medical Research Council. Variables independently associated with therapeutic inertia in patients with severe uncontrolled COPD (CAT score > 16 and at least one exacerbation in the past 3 months); *p < 0.01; **p < 0.0001. Control of COPD Perceived by Physicians In the 3479 patients with uncontrolled COPD (CAT score > 16 and at least one exacerbation in the past 3 months), control of COPD was rated as very poor or poor (ie, uncontrolled COPD) in 1105 patients, with a rate of agreement of 31.8%, whereas control of COPD was rated as satisfactory, good or very good (ie, controlled COPD) in the remaining 2373 patients, with a discordance rate of 68.2%. By contrast, among 1322 patients with controlled COPD (CAT score ≤ 16 and no exacerbations in the past 3 months), physicians reported the degree of control as very poor or poor in 83, with a discordance rate of 6.3% (Figure 2). Figure 2Control of COPD perceived by physicians rated on a 5-point Likert scale as very poor/poor (uncontrolled COPD) and satisfactory/good/very good (controlled COPD) and objective measures based on the CAT score and exacerbations in the past 3 months. Discrepancy rates of 68.2% for uncontrolled COPD and 6.3% for controlled COPD.The image contains two pie charts. The first chart represents 3479 patients with uncontrolled COPD, defined by a CAT score less than 16 and at least one exacerbation in the past 3 months. It shows 31.8 percent (1105 patients) with uncontrolled COPD and 68.2 percent (2373 patients) with controlled COPD. The second chart represents 1322 patients with controlled COPD, defined by a CAT score greater than or equal to 16 and no exacerbations in the past 3 months. It shows 6.3 percent (83 patients) with uncontrolled COPD and 93.7 percent (1239 patients) with controlled COPD.Two pie charts showing COPD control perceived by physicians, with percentages for controlled and uncontrolled cases. Control of COPD perceived by physicians rated on a 5-point Likert scale as very poor/poor (uncontrolled COPD) and satisfactory/good/very good (controlled COPD) and objective measures based on the CAT score and exacerbations in the past 3 months. Discrepancy rates of 68.2% for uncontrolled COPD and 6.3% for controlled COPD. As shown in Table 3, variables associated with uncontrolled COPD as compared to controlled COPD based on physicians’ criteria were current smoking, lower levels of physical activity, a lower percentage of patients being managed in secondary care, reduced family/social support, higher comorbidity (Charlson index), greater dyspnea (mMRC score), more exacerbations, higher CAT score, more severe airflow obstruction, yellow/green sputum, and poorer adherence to inhaled therapy. The patients’ perceived satisfaction with treatment was rated as poor/very poor in 41.6% of cases in the uncontrolled COPD group vs. 4.0% in the controlled group (p < 0.0001). Independent variables associated with physicians’ perception of controlled COPD in patients with uncontrolled disease are shown in Figure 3. The patients’ satisfaction with treatment was the most influential factor (OR 12.15, 95% CI 9.14–16.16) followed by the use of triple inhaled therapy (OR 2.27, 95% CI 1.46–3.53).Table 3Variables Associated with Physician’s Perception of Controlled and Uncontrolled COPDVariablesUncontrolled COPD (n = 1105)Controlled COPD (n = 2373)P valueSex, n (%) Males894 (80.9)1949 (82.1)0.383 Females211 (19.1)424 (17.9)Age, years, mean (SD)70.1 (9.8)69.9 (9.2)0.736Body mass index, kg/m2, mean (SD)27.2 (5.2)27.6 (5.2)0.439Smoking, n (%) Smoker325 (29.4)510 (21.5)< 0.001 Ex-smoker/never smoker780 (70.6)1863 (78.5)Years since COPD diagnosis, mean (SD)10.8 (7.5)9.7 (6.6)0.005Level of physical activity, n (%) High87 (7.9)335 (13.7)< 0.001 Moderate329 (29.8)982 (41.4) Low/inactive619 (56.0)916 (38.6) Missing70 (6.3)140 (5.9)Level of care, n (%) Primary104 (9.4)175 (7.4)< 0.001 Secondary (pulmonology, internal medicine)980 (88.7)2155 (90.8) Missing21 (1.9)43 (1.8)Working status, n (%) Active158 (14.3)312 (13.1)0.355 Inactive947 (85.7)2061 (86.9)Place of residence, n (%) Rural180 (16.3)404 (17.0)0.763 Semiurban222 (20.1)456 (19.2) Urban703 (63.6)1513 (63.8)Family/social support, n (%) Yes425 (38.5)761 (32.1)0.002 No680 (61.5)1612 (67.9)Comorbidities, n (%) Allergy/atopy99 (9.0)150 (6.3)0.005 Nasal polyposis20 (1.8)35 (1.5)0.463 Allergic rhinitis90 (8.1)142 (6.0)0.017Charlson comorbidity index, mean (SD)2.6 (1.8)2.1 (1.5)< 0.001Dyspnea mMRC scale, mean (SD)2.6 (0.9)2.2 (0.9)< 0.001Exacerbations in the last year, n (%) Severe (≥1)602 (54.5)905 (38.1)< 0.001 Moderate (≥2)515 (46.6)631 (26.6)< 0.001CAT score, mean (SD)23.7 (6.7)18.2 (7.0)< 0.001Post-bronchodilator FEV1% predicted, mean (SD)37.2 (8.6)38.9 (8.5)< 0.001Sputum color, n (%) White641 (58.0)1896 (80.0)< 0.001 Yellow/green464 (42.0)475 (20.0)Eosinophil blood count, cells/µL, n (%) <100208 (18.8)395 (16.6)0.080 100-300478 (43.3)1108 (46.7) >300139 (12.6)343 (14.5) Missing280 (25.3)527 (22.2)TAI adherence to inhaled therapy, n (%) Good493 (44.6)1424 (60.0)< 0.001 Intermediate209 (18.9)436 (18.4) Poor338 (30.6)388 (16.4) Missing65 (5.9)125 (5.3)Abbreviations: SD, standard deviation; mMRC, modified Medical Research Council; CAT, COPD assessment test; TAI, test of adherence inhalers. Figure 3Variables independently associated with the physicians’ perception of controlled COPD (5-point Likert scale rated as satisfactory/good/very good) in patients with uncontrolled COPD (COPD assessment test [CAT] score > 16 and at least one exacerbation in the past 3 months); *p < 0.01; **p < 0.0001.The graph shows odds ratios for various factors affecting the physicians’ perception of controlledChronic Obstructive Pulmonary Disease (COPD). The x-axis is labeled 'Odds ratio' ranging from 0 to 17. The y-axis lists factors such as home oxygen therapy, SABA treatment, triple therapy, dual inhaled therapy, Charlson comorbidity index, satisfaction with treatment, adherence to inhaled therapy, degree of dyspnea (mMRC), sputum color, CAT score, post-bronchodilation FEV1 percent predicted, moderate and severe exacerbations and smoking status. Each factor is associated with an odds ratio and a 95 percent confidence interval. Notable values include satisfaction with treatment (12.15, 9.14–16.16), triple therapy versus monotherapy (2.27, 1.46–3.53) and smoking status (1.88, 1.50–2.35). The graph highlights the significance of treatment satisfaction and therapy types in COPD management.A graph showing odds ratios for factors affecting the physicians’ perception of controlled COPD.Abbreviations: CI, confidence interval; SABA, short-acting beta2-agonist; mMRC, modified Medical Research Council; FEV1, forced expiratory volume in one second. Table 3Variables Associated with Physician’s Perception of Controlled and Uncontrolled COPDVariablesUncontrolled COPD (n = 1105)Controlled COPD (n = 2373)P valueSex, n (%) Males894 (80.9)1949 (82.1)0.383 Females211 (19.1)424 (17.9)Age, years, mean (SD)70.1 (9.8)69.9 (9.2)0.736Body mass index, kg/m2, mean (SD)27.2 (5.2)27.6 (5.2)0.439Smoking, n (%) Smoker325 (29.4)510 (21.5)< 0.001 Ex-smoker/never smoker780 (70.6)1863 (78.5)Years since COPD diagnosis, mean (SD)10.8 (7.5)9.7 (6.6)0.005Level of physical activity, n (%) High87 (7.9)335 (13.7)< 0.001 Moderate329 (29.8)982 (41.4) Low/inactive619 (56.0)916 (38.6) Missing70 (6.3)140 (5.9)Level of care, n (%) Primary104 (9.4)175 (7.4)< 0.001 Secondary (pulmonology, internal medicine)980 (88.7)2155 (90.8) Missing21 (1.9)43 (1.8)Working status, n (%) Active158 (14.3)312 (13.1)0.355 Inactive947 (85.7)2061 (86.9)Place of residence, n (%) Rural180 (16.3)404 (17.0)0.763 Semiurban222 (20.1)456 (19.2) Urban703 (63.6)1513 (63.8)Family/social support, n (%) Yes425 (38.5)761 (32.1)0.002 No680 (61.5)1612 (67.9)Comorbidities, n (%) Allergy/atopy99 (9.0)150 (6.3)0.005 Nasal polyposis20 (1.8)35 (1.5)0.463 Allergic rhinitis90 (8.1)142 (6.0)0.017Charlson comorbidity index, mean (SD)2.6 (1.8)2.1 (1.5)< 0.001Dyspnea mMRC scale, mean (SD)2.6 (0.9)2.2 (0.9)< 0.001Exacerbations in the last year, n (%) Severe (≥1)602 (54.5)905 (38.1)< 0.001 Moderate (≥2)515 (46.6)631 (26.6)< 0.001CAT score, mean (SD)23.7 (6.7)18.2 (7.0)< 0.001Post-bronchodilator FEV1% predicted, mean (SD)37.2 (8.6)38.9 (8.5)< 0.001Sputum color, n (%) White641 (58.0)1896 (80.0)< 0.001 Yellow/green464 (42.0)475 (20.0)Eosinophil blood count, cells/µL, n (%) <100208 (18.8)395 (16.6)0.080 100-300478 (43.3)1108 (46.7) >300139 (12.6)343 (14.5) Missing280 (25.3)527 (22.2)TAI adherence to inhaled therapy, n (%) Good493 (44.6)1424 (60.0)< 0.001 Intermediate209 (18.9)436 (18.4) Poor338 (30.6)388 (16.4) Missing65 (5.9)125 (5.3)Abbreviations: SD, standard deviation; mMRC, modified Medical Research Council; CAT, COPD assessment test; TAI, test of adherence inhalers. Variables Associated with Physician’s Perception of Controlled and Uncontrolled COPD Abbreviations: SD, standard deviation; mMRC, modified Medical Research Council; CAT, COPD assessment test; TAI, test of adherence inhalers. Variables independently associated with the physicians’ perception of controlled COPD (5-point Likert scale rated as satisfactory/good/very good) in patients with uncontrolled COPD (COPD assessment test [CAT] score > 16 and at least one exacerbation in the past 3 months); *p < 0.01; **p < 0.0001. Actions Taken During the Visit According to COPD Control Actions taken during the visit and modifications of pharmacological treatment were similar regardless of whether COPD control was assessed using objective methods or based on physicians’ perceptions (Table 4). Compared with controlled COPD, uncontrolled COPD was associated with significantly less frequent treatment withdrawal and dose reduction, but significantly more frequent non-pharmacological modifications and change of inhaler devices. Treatment addition was significantly more frequent in uncontrolled COPD but only when disease control was assessed by objective methods.Table 4Actions Taken During the Clinical Visit According to the Control of Disease in the Overall Study Population of 4801 Patients with Severe COPDActionsControl of COPD(CAT Score and Exacerbations in the Past 3 Months)Control of COPD Based on Physicians’ CriteriaUncontrolled (CAT Score < 16 and At Least 1 Exacerbation) (n = 3479)Controlled (CAT Score ≥ 16 and No Exacerbations) (n = 1322)P valueUncontrolled (n = 1105)Controlled (Discordant Perception) (n = 2373)P ValueWhere any actions taken during the visit? n (%) No1519 (43.7)840 (63.5)< 0.0001110 (10.0)1408 (59.3)< 0.0001 Yes1960 (56.3)482 (36.5)995 (90.0)965 (40.7)Modification of the prescribed medication, n (%)934 (47.7)162 (33.6)< 0.0001652 (65.5)282 (29.2)< 0.0001 Treatment addition538 (57.6)68 (42.0)0.0002386 (59.2)152 (53.9)0.1323 Treatment withdrawal52 (5.6)25 (15.4)< 0.000125 (3.8)27 (9.6)0.0004 Treatment substitution (switch)301 (32.2)55 (34.0)0.665221 (33.9)80 (28.4)0.0971 Dose increase112 (12.0)14 (8.6)0.21783 (12.7)29 (10.3)0.2907 Dose reduction18 (1.9)10 (6.2)0.0015 (0.8)13 (4.6)<0.0001Non-pharmacological modification, n (%)232 (11.8)24 (5.0)< 0.0001146 (14.7)86 (8.9)<0.0001Request for additional diagnostic tests, n (%)489 (24.9)119 (24.7)0.905229 (23.0)260 (26.9)0.0445Patient education, n (%)939 (47.9)250 (51.7)0.119405 (40.7)534 (55.3)<0.0001Change of inhaler device, n (%)232 (11.8)31 (6.4)0.0006171 (17.2)61 (6.3)<0.0001Other, n (%)151 (7.7)42 (8.7)0.46187 (8.7)64 (6.6)0.0797Abbreviation: CAT, COPD assessment test. Table 4Actions Taken During the Clinical Visit According to the Control of Disease in the Overall Study Population of 4801 Patients with Severe COPDActionsControl of COPD(CAT Score and Exacerbations in the Past 3 Months)Control of COPD Based on Physicians’ CriteriaUncontrolled (CAT Score < 16 and At Least 1 Exacerbation) (n = 3479)Controlled (CAT Score ≥ 16 and No Exacerbations) (n = 1322)P valueUncontrolled (n = 1105)Controlled (Discordant Perception) (n = 2373)P ValueWhere any actions taken during the visit? n (%) No1519 (43.7)840 (63.5)< 0.0001110 (10.0)1408 (59.3)< 0.0001 Yes1960 (56.3)482 (36.5)995 (90.0)965 (40.7)Modification of the prescribed medication, n (%)934 (47.7)162 (33.6)< 0.0001652 (65.5)282 (29.2)< 0.0001 Treatment addition538 (57.6)68 (42.0)0.0002386 (59.2)152 (53.9)0.1323 Treatment withdrawal52 (5.6)25 (15.4)< 0.000125 (3.8)27 (9.6)0.0004 Treatment substitution (switch)301 (32.2)55 (34.0)0.665221 (33.9)80 (28.4)0.0971 Dose increase112 (12.0)14 (8.6)0.21783 (12.7)29 (10.3)0.2907 Dose reduction18 (1.9)10 (6.2)0.0015 (0.8)13 (4.6)<0.0001Non-pharmacological modification, n (%)232 (11.8)24 (5.0)< 0.0001146 (14.7)86 (8.9)<0.0001Request for additional diagnostic tests, n (%)489 (24.9)119 (24.7)0.905229 (23.0)260 (26.9)0.0445Patient education, n (%)939 (47.9)250 (51.7)0.119405 (40.7)534 (55.3)<0.0001Change of inhaler device, n (%)232 (11.8)31 (6.4)0.0006171 (17.2)61 (6.3)<0.0001Other, n (%)151 (7.7)42 (8.7)0.46187 (8.7)64 (6.6)0.0797Abbreviation: CAT, COPD assessment test. Actions Taken During the Clinical Visit According to the Control of Disease in the Overall Study Population of 4801 Patients with Severe COPD Abbreviation: CAT, COPD assessment test. Discussion The present study carried out in the real-world setting in a large population of outpatients with severe uncontrolled COPD shows a high rate of TI of 43.7%. The gap between guideline recommendations and actual COPD management has a relevant effect on the progression of the disease, increasing the risk of exacerbation episodes, hospitalizations, airflow impairment, healthcare utilization, decreasing quality of life, and affecting morbidity and mortality.25,26 Recently, a definition of COPD control based on the concepts of impact and clinical stability, which can be measured by clinical changes (ie, absence of significant clinical worsening and absence of exacerbations in the past 3 months), CAT, degree of dyspnea, level of physical activity, sputum color, and the use of rescue medication has been proposed as a useful tool in the decision-making regarding changes in medication in the follow-up of COPD patients.27 It has been shown that evaluation of control status at each clinical visit provides relevant prognostic information about the risk of exacerbation in the next 6 months.28 In our study, objective assessment of COPD control was based on a CAT score cutoff of ≤ and > 16 together with history of exacerbations within the past 3 months, a criteria that has been previously used in the CLAVE study.19 Interestingly, patients’ level of satisfaction with treatment emerged as the most influential variable associated with both TI and the physicians’ perception of disease control. These findings suggest that physicians may be less likely to modify treatment regimens when patients express satisfaction with their current treatment. We also identified smoking status, use of SABA, adherence to inhaled therapy, using dual or triple therapy, absence of severe exacerbations in the previous year, and specialized level of care as independent factors related to TI. These findings are in agreement with data of the EPOCONSUL audit study,18 in which being an ex-smoker, lower number of exacerbations in the previous year, and dual inhaled therapy were significant factors associated with TI. Insufficient time, unawareness of guidelines, failure to address comorbidities, lack of spirometry use, and a predominantly reactive rather than proactive attitude have been reported as major risk factors contributing to TI among healthcare providers.29 In relation to the influence of the healthcare setting (primary care vs. specialized care), management of patients by primary care physicians or specialists in internal medicine as compared with pneumologists were significant factors associated with TI. In a survey study of 784 practicing primary care physicians, although 55% were aware of major COPD guidelines, only 25% used them to guide decision-making, with guideline users being more likely to indicate spirometry and initiate therapy for mild respiratory symptoms.30 Although primary care plays a key role in the diagnosis, management, and follow-up of COPD patients, other studies have shown suboptimal adherence to clinical guidelines by family physicians, highlighting the need to improve the quality of care of COPD in this setting.31,32 Discrepancies between physicians’ perception of COPD control showed marked differences according to controlled or controlled COPD based on objective measures. Among patients with uncontrolled COPD, 68.2% of physicians reported good control, but among patients with controlled COPD, only 6.3% reported poor control. The high proportion of physicians perceiving disease control as adequate in patients with severe uncontrolled COPD underscores the need to address this issue, as uncontrolled patients are at an increased risk of moderate and, particularly, severe exacerbations, which are known to contribute to lung function decline, reduced quality of life, and increase morbidity and mortality.33,34 The 2021 EPOCONSUL audit, which evaluated practice in pulmonology clinics, revealed that control was assessed in only a quarter of the visits audited.18 A quantitative score has recently been developed and validated as a predictor of the risk of complications, allowing clinicians to more accurately identify patients with insufficient control.35,36 This facilitates clinical decision-making, treatment adjustments, and monitoring of changes over time. In the multivariate analysis, besides the influence of patients’ satisfaction with treatment, the absence of severe exacerbations in the previous year was an independent factor associated with discrepancies between COPD control perceived by physicians and evaluated objectively. It has been estimated that the number of underreported exacerbations may be 2 to 5 times higher than that of reported ones, but such events exert a similar negative impact on the health status of patients with COPD to those that are reported.37 Moreover, early identification and quickly starting treatment can result in prompt recovery, improving patient outcomes and reducing healthcare costs.38 It has been shown that a single moderate exacerbation in the previous year significantly increases the probability of subsequent exacerbations and respiratory hospitalizations39 and even mortality40 during the following years. It is crucial to avoid TI in these patients who have had only one moderate exacerbation in the previous year and should lead to an escalation therapy.41 In this respect, in patients with history of one moderate exacerbation in the previous year who are not controlled with single or dual maintenance therapy, switching to triple therapy reduces moderate and severe exacerbations, and improves pulmonary function and quality of life.42 The fact that healthcare professionals in all healthcare settings should be able to recognize COPD exacerbations, make appropriate referrals, document episodes accurately in medical records, and support patients to recognize and report exacerbations, has been emphasized as one of the actions points to reduce TI in COPD.15 Introduction of modifications in the pharmacological treatment depended on the control of COPD, with less frequent changes among patients with controlled COPD. As may be expected, treatment withdrawal and dose reduction were more frequently recorded in patients with controlled COPD, whereas the addition of pharmacological agents was more common in patients with uncontrolled disease. Modification of non-pharmacological measures especially changes of the inhaler device was also more common in uncontrolled COPD. Among control of COPD according to the physicians’ perceptions, patient education and request for additional diagnostic tests were significantly more frequent in patients with uncontrolled COPD. However, it should be noted that in 43.7% of patients with uncontrolled COPD based on objective methods, no actions were undertaken. This is a clinically relevant finding and is consistent with the real-world challenge of COPD undertreatment. In the Copenhagen General Population Study aimed to characterize undertreatment in COPD, only 30% of 920 individuals with an FEV1 < 60% predicted were treated with medication in the year before the examination increasing to 42.2% the year after.43 Respiratory infections were the strongest predictor of treatment with medications, which indicates that most COPD treatment is initiated because of acute exacerbations.43 Underdiagnosis, however, has been recognized as an important cause of undertreatment, leading to patients reaching advanced disease stages without receiving appropriate treatment. Early detection of COPD is a crucial strategy for improving the outcome of patients and reducing the burden on healthcare systems.44 The present results should be interpreted considering the post hoc design and the exploratory nature of the analysis. However, assessment of TI in a large cohort of severe uncontrolled COPD patients from a real-world setting and evaluation of discrepancies between physicians’ perception of disease control and controlled determined by objective measures (CAT score and exacerbations) are novel aspects of the study. Conclusions In a post hoc analysis of a large population of outpatients with severe uncontrolled COPD attended in real-world clinical practice, there was a high rate of TI of 43.7%. Discrepancies between physicians’ perception of COPD control revealed marked differences according to controlled or uncontrolled COP based on objective measures (CAT score and exacerbations in the past 3 months). Notably, among patients with uncontrolled COPD, 68.2% of physicians reported good control of the disease. This discordance between physicians’ perception of the control of COPD and results of objective measures, together with a high rate of TI have important clinical implications for practice. A multifaceted approach is needed to have tools available (eg, protocols, decision tools) to facilitate the standardization of the assessment of the degree of control in the follow-up of patients with COPD. In this respect, targeting factors related to healthcare professionals, patients, and health systems would ensure the development of protocol-driven actions to improve a patient-centered approach over the clinical course of the patients.