Advertisement
Research Article| Volume 154, P12-17, July 2019

Triple inhaled therapy in COPD patients: determinants of prescription in primary care

Open ArchivePublished:May 29, 2019DOI:https://doi.org/10.1016/j.rmed.2019.05.022

      Highlights

      • One fifth of primary care patients with COPD are prescribed with triple inhaled therapy over 4.5 years.
      • Older age, cigarette smoking, COPD severity and heart failure trigger the escalation to the triple inhaled therapy.
      • The knowledge of such factors may help GPs and specialist to better personalize pharmacological treatment in COPD patients.

      Abstract

      Objective

      To assess the incidence and determinants of the triple inhaled therapy in chronic obstructive pulmonary disease (COPD) primary care patients.

      Methods

      Data derived from the Health Search Database (HSD) gathering information on 700 Italian general practitioners. A cohort of COPD patients, prescribed for the first time with inhaled treatments, was followed-up between January 2002 and December 2014. The outcome was the first incident prescription of a triple inhaled therapy, namely the combination of inhaled corticosteroids (ICS), long-acting beta agonists (LABA), and long-acting muscarinic antagonists (LAMA). Cox regressions were used to test the association (hazard ratios, HR) between candidate determinants and the outcome.

      Results

      Out of 17589 patients (mean age 71.1 ± 11.3 years; 37.4% females), 3693 (21%) were prescribed with a triple inhaled therapy during follow-up. Older age (HR = 1.79 to 2.61), current and former smoking habit (HR = 1.72 and 1.66), higher GOLD stage (HR = 1.45 to 2.79), the number of moderate and severe COPD exacerbations (HR = 1.10 to 2.63), and heart failure (HR = 1.17) resulted statistically significantly associated with an increased incident prescription of the triple inhaled therapy. Female sex (HR = 0.80) and some comorbidities (HR = 0.21 to 0.87) resulted negatively associated with the outcome. Furthermore, patients initially treated with LAMA (HR = 1.5) and LABA/ICS (HR = 1.23) were more likely to escalate to the triple therapy, than those on LABA. Conversely, patients initially treated with ICS presented a negative hazard (HR = 0.72).

      Conclusions

      The knowledge of demographic and clinical determinants of the escalation to the triple inhaled therapy in real-world COPD patients may help clinicians to better personalize respiratory pharmacological treatments of their patients, and inform international societies that issue clinical guidelines.

      Keywords

      1. What is already known on the subject

      Real-world COPD triple inhaled therapy prescription does not always reflect guidelines. Which factors trigger the escalation to triple therapy in primary care COPD patients are not clear.

      2. What this paper adds

      We showed that 21% of primary care patients with COPD are prescribed with triple inhaled therapy over a period of 4.5 years. Older age, cigarette smoking, COPD severity and heart failure make this escalation more likely to happen. The knowledge of such factors may help general practitioner and specialist to better personalize pharmacological treatment in COPD patients.

      3. Introduction

      Chronic obstructive pulmonary disease (COPD) is a prevalent and burdensome condition that affects 8–15% of the general population, with figures increasing in older age [
      • Adeloye D.
      • Chua S.
      • Lee C.
      • Basquill C.
      • Papana A.
      • Theodoratou E.
      • Nair H.
      • Gasevic D.
      • Sridhar D.
      • Campbell H.
      • Chan K.Y.
      • Sheikh A.
      • Rudan I.
      Global Health Epidemiology Reference Group (GHERG)
      Global and regional estimates of COPD prevalence: systematic review and meta-analysis.
      ]. COPD represents the fourth major contributor to the number of years of life spent with disability in advanced age, and the fourth cause of death in high-income countries, candidate to occupy soon the third place [
      • Murray C.J.
      • Lopez A.D.
      Alternative projections of mortality and disability by cause 1990-2020: global burden of disease study.
      ]. During the last two decades, relevant improvements in pharmacological treatment have guaranteed better symptoms control to COPD patients, with relevant impact on their health and perceived quality of life. Namely, several randomized controlled trials (RCT) have shown that the association of inhaled corticosteroids (ICS) with a long-acting beta agonist (LABA) and with a long-acting muscarinic antagonist (LAMA) results in a reduction in the number of exacerbations, especially among those with severe obstruction [
      • Calverley P.M.
      • Anderson J.A.
      • Celli B.
      • et al.
      Salmeterol and fluticasone propionate and survival in chronic obstructive pulmonary disease.
      ,
      • Tashkin D.P.
      • Celli B.
      • Senn S.
      • et al.
      A 4-year trial of tiotropium in chronic obstructive pulmonary disease.
      ,
      • Mahler D.A.
      • Donohue J.F.
      • Barbee R.A.
      • et al.
      Efficacy of salmeterol xinafoate in the treatment of COPD.
      ].
      Since 2007, the Global Initiative for Obstructive Lung Disease (GOLD) recommendations have suggested the addition of an ICS to the inhaled bronchodilator therapy in patients with severe COPD and frequent exacerbations [
      • Gold P.M.
      The 2007 GOLD Guidelines: a comprehensive care framework.
      ]. However, these same guidelines report weak evidence regarding the criteria to be followed in stepping-up to the triple inhaled therapy [
      • Rodriguez-Roisin R.
      • Rabe K.F.
      • Vestbo J.
      • Vogelmeier C.
      • Agusti A.
      Global initiative for chronic obstructive lung disease (GOLD) 20th anniversary: a brief history of time.
      ]. Notably, recent RCTs have reported significant exacerbations reduction in COPD patients randomized to receive a triple therapy LABA-LAMA-ICS via a single inhaler, as compared with patients randomized to the combination LABA-ICS or LAMA-ICS [
      • Lipson D.A.
      • Barnhart F.
      • Brealey N.
      • et al.
      Once-Daily single-inhaler triple versus dual therapy in patients with COPD.
      ,
      • Singh D.
      • Papi A.
      • Corradi M.
      • et al.
      Single inhaler triple therapy versus inhaled corticosteroid plus long-acting beta2-agonist therapy for chronic obstructive pulmonary disease (TRILOGY): a double-blind, parallel group, randomised controlled trial.
      ]. However, despite clear indications from international and national guidelines, the analysis of real-life prescribing patterns has shown significant under- and over-prescription in COPD patients, in particular with regard to the triple inhaled therapy. For example, a study carried out in the UK, reported an excessive prescription of the triple combination in COPD patients with low risk of exacerbations, with most patients receiving ICS, irrespectively of severity of airflow limitation, asthma diagnosis, and exacerbation history [
      • Price D.
      • West D.
      • Brusselle G.
      • et al.
      Management of COPD in the UK primary-care setting: an analysis of real-life prescribing patterns.
      ]. An abused prescription of ICS raises several concerns about safety, particularly regarding the risk of pneumonia [
      • Ernst P.
      • Saad N.
      • Suissa S.
      Inhaled corticosteroids in COPD: the clinical evidence.
      ]. In general, inadequate prescription may affect the efficacy/safety profile of COPD drugs, likely influencing treatment adherence, triggered by the patients’ perception of scarce benefit. On the other hand, understanding the reasons why COPD patients are prescribed with specific combinations of COPD drugs, as for example the triple inhaled therapy, may help improving both the management and the adherence to the treatment. For this reason, identifying the profile of real-life patients prescribed with specific classes of drugs, sounds relevant to improve the quality of prescription.
      In a recent Italian study [
      • DiMarco F.
      • Santus P.
      • Terraneo S.
      • et al.
      Characteristics of newly diagnosed COPD patients treated with triple inhaled therapy by general practitioners: a real world Italian study.
      ], conducted in primary care, the authors reported the incidence of triple inhaled therapy was as high as 6% among patients newly (i.e. 1 year) diagnosed with COPD. They also identified some potential determinants of this transition, among which age, sex, and a few clinical conditions. However, some relevant information has not been considered in this study, among which the impact of smoking, COPD stage and the number and entity of disease exacerbations, leaving open several avenues of investigation. The aim of the present study was to assess the incidence rate and the determinants of the triple inhaled therapy in a large cohort of COPD primary care patients.

      4. Methods

      4.1 Data source

      For the present study we analyzed data from the Health Search Database (HSD). HSD is an Italian general practice database used for research purposes, that gathers a significant deal of information, including patients' demographics, clinical diagnoses, drug prescriptions, specialist referrals, and date of death. Clinical examinations, drug prescriptions and diseases are coded in accordance with the National Health code system, the Anatomical Therapeutic Chemical (ATC) classification system and the International Classification of Diseases 9th Revision Clinical Modification (ICD-9CM), respectively. Out of a network of about 1000 general practitioners (GPs) – which register patients’ data on a voluntary basis – we selected those 700 GPs who fulfill the up-to standard quality criteria for data registration. Such practitioners are homogeneously distributed across Italy and cover a population of more than one million individuals. The representativeness and reliability of HSD data for epidemiological research are supported by a number of studies [
      • Vetrano D.L.
      • Bianchini E.
      • Onder G.
      • et al.
      Poor adherence to chronic obstructive pulmonary disease medications in primary care: role of age, disease burden and polypharmacy.
      ]. According to a by-law on the classification and implementation of observational drug-related research, as issued by the Italian Medicines Agency, the present study does not require approval by an ethics committee. The results of the present study have been reported in keeping with the STROBE recommendations.

      4.2 Cohorts definition

      To address the aim of the present study, a cohort of 17589 patients was derived from HSD within the time frame 1 January 2002 and 31 December 2014. Data from patients 40 years or older with a diagnosis of COPD (ICD-9CM 491.2* or 496*) were inspected. We selected COPD patients with an incident prescription of LAMA, LABA or ICS, as single therapy or double association, and with a coverage of at least one month. The date of the first prescription of such drugs will be chosen as the index date. Patients with a triple prescription LABA-LAMA-ICS were excluded from the cohort.

      4.3 Outcomes definition

      The incident prescription of the free triple inhaled therapy LABA-LAMA-ICS, with a coverage of at least one month, was considered the outcome of the present study. The date of the first day of the month of concurrent use was considered the event date. Patients were followed until either the development of the outcome, death, loss of contact with the GP or the end of the study (31 December 2014).

      4.4 Candidate determinants

      Demographic, lifestyle and clinical information was collected until the index date. Airflow limitation level was staged according to the GOLD 2007 recommendations. When spirometry data were available, the last measurement, expressed as % of the predicted first-second forced expiratory volume (FEV1), preceding or on the index date was adopted to classify airflow obstruction as mild (GOLD I), moderate (GOLD II), severe (GOLD III) or very severe (GOLD IV). In this study, a moderate COPD exacerbation was defined as the incident co-prescription, in the same day (±3 days) of an antibiotic (ATC J01*) and a corticosteroid drug (ATC H02*). A severe COPD exacerbation was defined when a hospitalization due to COPD occurred. The number of moderate and severe exacerbations within the year prior to the index date were considered, as well as the number of GP visits triggered by a respiratory symptomatology. The last available records preceding or corresponding to the index date was used to define smoking habit. Obesity was defined as a body mass index≥30 kg/m2 or when the ICD-9CM code 278.0 was recorded within the last year preceding or on the index date; concurrent chronic diseases were identified through their ICD-9CM recorded in all the available period preceding or on the index date. Polypharmacy was defined as the contemporary use of ≥5 ATCs within 6 months preceding or on the index date.

      4.5 Analytical approach

      Baseline patients' characteristics were reported as mean ± standard deviation (SD) or absolute numbers and percentage (%), as appropriate. The association (hazard ratio [HR] and 95% confidence interval [CI]) between incident triple inhaled therapy and potential determinants was tested through Cox regression models. The candidate determinants entered the model or were excluded (p value equal to 0.10 or 0.15 for entering or exiting variables, respectively) in accordance with a stepwise procedure. The variable related to COPD severity were a priori retained into the model. Dummy categories were built for missing values. The proportional hazards assumption was assessed by regressing the scaled Schoenfeld's residuals against the incident outcome. No violation of proportionality was detected. The following sensitivity analyses have been carried out to test the robustness of the results: 1) in order to buffer the potential selection bias derived by the applied selection criteria, the analysis was repeated including also the prevalent users of at least one COPD inhaler therapy, namely, those that were already on treatment before the index date; 2) in order to test the accuracy of the outcome definition, the Cox regression was repeated considering as outcome the incident prescription of the triple inhaled therapy LABA-LAMA-ICS, with a coverage of at least two and three – instead of one – months; 3) in order to verify the burden of missing values for spirometry measures, the model was re-run selecting only those participants with at least one valid spirometry measure. All the analyses were carried out with Stata 13.0 (StataCorp).

      5. Results

      5.1 Determinants of incident triple inhaled therapy

      Out of 17589 COPD patients (mean age 71.1 ± 11.3 years; 37.4% females) included in the present study, 3693 (21%) were prescribed with a triple inhaled therapy during follow-up (mean follow-up 4.5 years). Among them, 3301 (89.4%) reached the triple therapy with a combination of LABA/ICS (in association) and LAMA, and 392 (10.6%) with a combination of LABA, LAMA and ICS. Patients that were prescribed with a LAMA at baseline took on average 4.2 years to develop the outcome. Those prescribed with a LABA took on average 2.3 years, those prescribed with an ICS took 4.2 years and those prescribed with a combination LABA/ICS took 3.7 years.
      Table 1 reports the baseline characteristics of the study population and the results of the univariate and multivariate models for the association between the incident triple inhaled therapy and its potential determinants. According to the multivariate analysis, females presented with a 20% lower likelihood of developing the outcome than males. Older age (with an attenuation after 80 years), being current or former smoker, a more severe GOLD stage, and the history of previous moderate and severe exacerbations, were associated with a higher likelihood of receiving the triple inhaled therapy. Similarly, there was a positive association in patients with a diagnosis of heart failure. On the contrary, several comorbidities, namely diabetes, hemiplegia/paraplegia, kidney disease, metastatic cancer and dementia were associated with a trend of lower likelihood of incident triple inhaled therapy. As compared with patients prescribed with a LABA at baseline, those initially prescribed with a LAMA and a combination LABA/ICS showed a 50% and 23% higher likelihood to initiate a triple inhaled therapy, and those initially prescribed with solely an ICS presented a 28% lower likelihood to initiate the triple therapy.
      Table 1Predictors (frequencies and associations) of incident triple inhaled therapy in COPD patients incident users of one or two COPD inhaled treatments at baseline.
      PredictorsN (%)UnivariateMultivariate
      HR (95%CI)HR (95%CI)
      Sex (ref. Males)
      Females6581 (37.4)0.70 (0.65–0.75)0.80 (0.75–0.87)
      Age groups (ref. 40–49)
      50-592012 (11.4)1.88 (1.5–2.35)1.79 (1.43–2.24)
      60-694474 (25.4)2.67 (2.17–3.3)2.5 (2.02–3.09)
      70-795989 (34.1)2.76 (2.24–3.4)2.61 (2.12–3.23)
      >=804339 (24.7)2.31 (1.86–2.88)2.33 (1.87–2.91)
      Smoking habit (ref. No)
      Current4248 (24.2)1.87 (1.64–2.13)1.72 (1.50–1.98)
      Former2898 (16.5)2.06 (1.79–2.36)1.66 (1.44–1.92)
      Obesity (ref. No)
      Yes1418 (8.1)1.05 (0.92–1.21)
      GOLD staging (ref. 1)
      2690 (3.9)1.68 (1.23–2.28)1.45 (1.07–1.98)
      3293 (1.7)2.41 (1.73–3.37)2.23 (1.60–3.11)
      472 (0.4)2.73 (1.70–4.39)2.79 (1.73–4.48)
      Mild exacerbations (ref. None)
      1-22817 (16)1.03 (0.94–1.12)1.10 (1.01–1.2)
      >2767 (4.4)1.01 (0.86–1.19)1.18 (1.03–1.39)
      Severe exacerbations (ref. None)
      1-22097 (11.9)2.77 (2.57–2.98)2.63 (2.44–2.84)
      Visit to GP's office due to respiratory symptoms (ref. None)
      >=11924 (10.9)0.98 (0.87–1.09)
      Starting therapy (ref. LABA)
      LAMA2456 (14.0)1.45 (1.27–1.66)1.50 (1.31–1.72)
      ICS7630 (43.4)0.66 (0.58–0.75)0.72 (0.64–0.82)
      LABA/ICS6032 (34.3)1.19 (1.05–1.34)1.23 (1.09–1.39)
      Comorbidities (ref. Absence of disease)
      Eosinophilia1508 (8.6)0.90 (0.78–1.03)
      Heart failure1143 (6.5)1.26 (1.10–1.46)1.17 (1.01–1.35)
      Coronary heart disease2780 (15.8)1.04 (0.95–1.15)
      Peripheral artery disease2200 (12.5)1.10 (0.99–1.21)
      Cerebrovascular disease2300 (13.1)0.98 (0.88–1.09)
      Arrhythmia2140 (12.2)1.01 (0.91–1.13)
      Asthma583 (3.3)0.96 (0.80–1.14)
      Other chronic pulmonary diseases1506 (8.6)1.03 (0.91–1.17)
      Reumatic disorders296 (1.7)0.83 (0.62–1.12)
      Peptic ulcer1413 (8)1.05 (0.94–1.19)
      Uncomplicated liver diseases333 (1.9)0.89 (0.68–1.16)
      Complicated liver diseases31 (0.2)1.08 (0.45–2.60)
      Uncomplicated diabetes3073 (17.5)0.99 (0.91–1.08)
      Complicated diabetes16 (0.1)0.32 (0.05–2.28)0.23 (0.03–1.65)
      Hemiplegia/paraplegia62 (0.4)0.53 (0.24–1.19)0.50 (0.22–1.12)
      Kidney disease7007 (39.8)0.91 (0.85–0.97)0.87 (0.81–0.94)
      Cancer1913 (10.9)1.11 (0.99–1.23)
      Metastatic cancer44 (0.3)0.22 (0.03–1.54)0.21 (0.03–1.49)
      Depression2375 (13.5)0.96 (0.86–1.06)
      Dementia211 (1.2)0.62 (0.39–1.00)0.69 (0.43–1.11)
      AIDS/HIV13 (0.1)0.39 (0.06–2.79)
      Polypharmacy13057 (74.2)0.95 (0.88–1.02)
      Figures are reported as absolute number and proportion (%). HR = hazard ratio; 95%CI = 95% confidence interval. Missing values: smoking habit 8061, obesity 13356, GOLD stage 16182.
      Table 2 shows the results of the first analysis, extended to patients that were prevalent users of one or two COPD inhaled drugs at baseline (N = 27572). Among them, 6851 (24.8%) developed the outcome, with the results of the main analysis on the predictors being mostly confirmed.
      Table 2Predictors (frequencies and associations) of incident triple inhaled therapy in COPD patients prevalent users of one or two COPD inhaled treatments at baseline.
      PredictorsN (%)UnivariateMultivariate
      HR (95%CI)HR (95%CI)
      Sex (ref. Males)
      Females10404 (37.7)0.68 (0.65–0.72)0.76 (0.73–0.81)
      Age groups (ref. 40–49)
      50-593006 (10.9)1.76 (1.51–2.06)1.76 (1.51–2.06)
      60-696960 (25.2)2.32 (2.01–2.69)2.32 (2.00–2.69)
      70-799942 (36.1)2.34 (2.03–2.71)2.38 (2.05–2.76)
      >=806512 (23.6)1.84 (1.58–2.15)2.01 (1.72–2.36)
      Smoking habit (ref. No)
      Current5650 (20.5)1.85 (1.68–2.03)1.76 (1.59–1.94)
      Former4664 (16.9)2.01 (1.82–2.21)1.64 (1.48–1.81)
      Obesity (ref. No)
      Yes2260 (8.2)0.96 (0.87–1.06)
      GOLD staging (ref. 1)
      2827 (3.0)1.66 (1.27–2.17)1.54 (1.18–2.01)
      3350 (1.3)2.51 (1.88–3.36)2.38 (1.78–3.18)
      481 (0.3)2.52 (1.63–3.89)2.32 (1.50–3.59)
      Mild exacerbations (ref. None)
      1-25201 (18.9)1.16 (1.10–1.24)1.15 (1.08–1.22)
      >21767 (6.4)1.33 (1.22–1.46)1.28 (1.17–1.40)
      Severe exacerbations (ref. None)
      1-24172 (15.1)2.54 (2.41–2.67)2.4 (2.27–2.53)
      Visit to GP's office due to respiratory symptoms (ref. None)
      >=12634 (9.6)0.95 (0.87–1.04)
      Starting therapy (ref. LABA)
      LAMA2735 (9.9)1.24 (1.12–1.37)
      ICS13250 (48.1)0.65 (0.60–0.70)
      LABA/ICS8816 (32.0)1.04 (0.96–1.12)
      Comorbidities (ref. Absence of disease)
      Eosinophilia1826 (6.6)0.86 (0.76–0.97)
      Heart failure1690 (6.1)1.11 (0.99–1.24)
      Coronary heart disease4281 (15.5)1.04 (0.97–1.11)
      Peripheral artery disease3285 (11.9)1.07 (0.99–1.16)
      Cerebrovascular disease3309 (12.0)0.92 (0.85–1.00)0.90 (0.83–0.99)
      Arrhythmia3214 (11.7)0.94 (0.87–1.02)
      Asthma1252 (4.5)1.13 (1.02–1.25)1.24 (1.11–1.37)
      Other chronic pulmonary diseases2052 (7.4)1.04 (0.95–1.15)
      Reumatic disorders458 (1.7)0.74 (0.59–0.92)
      Peptic ulcer2309 (8.4)1.10 (1.01–1.20)
      Uncomplicated liver diseases562 (2.0)0.84 (0.70–1.02)
      Complicated liver diseases42 (0.2)1.00 (0.48–2.10)
      Uncomplicated diabetes4547 (16.5)0.94 (0.88–1.00)0.94 (0.88–1.01)
      Complicated diabetes30 (0.1)0.40 (0.13–1.24)0.38 (0.12–1.18)
      Hemiplegia/paraplegia83 (0.3)0.48 (0.24–0.95)0.42 (0.21–0.84)
      Kidney disease10338 (37.5)0.88 (0.84–0.93)0.90 (0.85–0.95)
      Cancer2965 (10.8)1.08 (1.00–1.18)
      Metastatic cancer55 (0.2)0.16 (0.02–1.13)0.17 (0.02–1.21)
      Depression3499 (12.7)0.94 (0.87–1.01)
      Dementia304 (1.1)0.54 (0.36–0.82)0.60 (0.40–0.91)
      AIDS/HIV15 (0.1)0.72 (0.18–2.88)
      Polypharmacy21235 (77.0)1.02 (0.97–1.08)
      Figures are reported as absolute number and proportion (%). HR = hazard ratio; 95%CI = 95% confidence interval. Missing values: smoking habit 13461, obesity 20823, GOLD stage 25883.
      The further sensitivity analyses showed results consistent with the main analysis (Tables S1, S2 and S3).

      6. Discussion

      Our findings show that one fifth of COPD primary care patients were prescribed with a combination of LABA, LAMA, and ICS within an average of 4.5 years after the prescription of the first COPD treatment. History of COPD exacerbations (both mild and severe), current or former smoking habit, older age, history of heart failure, and higher GOLD stage at baseline resulted strong determinants for the prescription of triple therapy. Our study is the first comprehensively inspecting the incidence of triple inhaled therapy and its related determinants in Italian patients affected by COPD.
      The combination of ICS, LAMA and LABA has shown beneficial effect on lung function, health status, symptoms control and on frequency of exacerbations [
      • Vogelmeier C.F.
      • Criner G.J.
      • Martinez F.J.
      • et al.
      Global strategy for the diagnosis, management, and prevention of chronic obstructive lung disease 2017 report. GOLD executive summary.
      ]. According to a large study carried out in the UK, triple inhaled therapy prescriptions sharply increased since 2002 when LAMA were available for the first time [
      • Bloom C.I.
      • Elkin S.L.
      • Quint J.K.
      Changes in COPD inhaler prescriptions in the United Kingdom, 2000 to 2016.
      ]. In 2007, the GOLD guidelines recommended the addition of an ICS to bronchodilator therapy in patients with severe COPD (FEV1 < 50% of predicted value) and frequent exacerbations. In accordance with our results, having had at least one exacerbation (mild or moderate) before the baseline, and having more severe COPD were the two strongest determinants (i.e.: highest hazards) of the escalation to the triple therapy, as already showed by Hurst JR et al. [
      • Hurst J.R.
      • Dilleen M.
      • Morris K.
      • Hills S.
      • Emir B.
      • Jones R.
      Factors influencing treatment escalation from long-acting muscarinic antagonist monotherapy to triple therapy in patients with COPD: a retrospective THIN-database analysis.
      ]. This finding is in line with what showed by Bloom et al. in a large study conducted in the UK [
      • Bloom C.I.
      • Elkin S.L.
      • Quint J.K.
      Changes in COPD inhaler prescriptions in the United Kingdom, 2000 to 2016.
      ]. We also found a different risk of escalation to triple therapy according to baseline inhaled therapy. Interestingly, both patients on LAMA and on LABA/ICS exhibited a higher risk of progression in comparison with those on LABA and ICS at baseline. This result was already showed in similar real-life study carried out in the UK [
      • Brusselle G.
      • Price D.
      • Gruffydd-Jones K.
      • et al.
      The inevitable drift to triple therapy in COPD: an analysis of prescribing pathways in the UK.
      ]. It is possible that part of this result is explained by differences in the population initially treated with LABA and those treated using other drugs at baseline, identifying different severity of the disease itself [
      • Decramer M.L.
      • Chapman K.R.
      • Dahl R.
      • et al.
      Once-daily indacaterol versus tiotropium for patients with severe chronic obstructive pulmonary disease (INVIGORATE): a randomised, blinded, parallel-group study.
      ,
      • Nannini L.J.
      • Poole P.
      • Milan S.J.
      • Holmes R.
      • Normansell R.
      Combined corticosteroid and long-acting beta(2)-agonist in one inhaler versus placebo for chronic obstructive pulmonary disease.
      ]. Our study also found that heart failure was associated with an increased risk of progression to triple therapy, supporting the link between cardiovascular diseases and the prescription of triple inhaled therapy as starting treatment in COPD, already showed by Di Marco [
      • DiMarco F.
      • Santus P.
      • Terraneo S.
      • et al.
      Characteristics of newly diagnosed COPD patients treated with triple inhaled therapy by general practitioners: a real world Italian study.
      ]. Our findings may be explained by the challenge posed by the differential diagnosis of dyspnea in patients affected by both heart failure and COPD, and by the mutual symptoms exacerbation of these two conditions. In such cases, a proper clinical and instrumental distinction might help to limit complex pharmacological regimens in these multimorbid patients [
      • Hawkins N.M.
      • Virani S.
      • Ceconi C.
      Heart failure and chronic obstructive pulmonary disease: the challenges facing physicians and health services.
      ]. Despite the fact older age was associated with an increased prescription of triple therapy for COPD, conditions associated with higher vulnerability (i.e. polypharmacy, presence of dementia, chronic kidney disease, solid tumors, hemi-paraplegia) were linked to a general trend of lower risk of progression toward triple therapy. This can be explained by a more cautious approach adopted by prescribers in presence of complex patients, where the further increase of the pharmacological burden might be associated with an increased risk of adverse events.
      Our results show that the incidence of triple therapy prescription in Italy is similar to the one found by Simeone JC [
      • Simeone J.C.
      • Luthra R.
      • Kaila S.
      • et al.
      Initiation of triple therapy maintenance treatment among patients with COPD in the US.
      ] and Mapel D [
      • Mapel D.
      • Laliberte F.
      • Roberts M.H.
      • et al.
      A retrospective study to assess clinical characteristics and time to initiation of open-triple therapy among patients with chronic obstructive pulmonary disease, newly established on long-acting mono- or combination therapy.
      ] among American patients, but lower to the one found in a recent study that inspected COPD therapy patterns in Sweden [
      • Sundh J.
      • Aberg J.
      • Hasselgren M.
      • et al.
      Factors influencing pharmacological treatment in COPD: a comparison of 2005 and 2014.
      ]. In contrast to the work of Sundh J et al., we found that female sex was negatively associated with the prescription of triple therapy. This result has been already reported in some studies [
      • DiMarco F.
      • Santus P.
      • Terraneo S.
      • et al.
      Characteristics of newly diagnosed COPD patients treated with triple inhaled therapy by general practitioners: a real world Italian study.
      ,
      • Souliotis K.
      • Kani C.
      • Papageorgiou M.
      • Lionis D.
      • Gourgoulianis K.
      Using big data to assess prescribing patterns in Greece: the case of chronic obstructive pulmonary disease.
      ,
      • Carrasco-Garrido P.
      • de Miguel-Diez J.
      • Rejas-Gutierrez J.
      • et al.
      Characteristics of chronic obstructive pulmonary disease in Spain from a gender perspective.
      ]. A lower awareness of COPD prevalence, symptoms and mortality in women among physicians have been suggested, but this topic is still debated [
      • Sundh J.
      • Aberg J.
      • Hasselgren M.
      • et al.
      Factors influencing pharmacological treatment in COPD: a comparison of 2005 and 2014.
      ,
      • Tsiligianni I.
      • Rodriguez M.R.
      • Lisspers K.
      • LeeTan T.
      • Infantino A.
      Call to action: improving primary care for women with COPD.
      ].
      This real-life study shows important results – besides the main ones – regarding the general therapeutic and diagnostic approach to primary care COPD patients in Italy. Interestingly, the most commonly prescribed starting therapy was ICS monotherapy (43.4%). While previous studies [
      • Hurst J.R.
      • Dilleen M.
      • Morris K.
      • Hills S.
      • Emir B.
      • Jones R.
      Factors influencing treatment escalation from long-acting muscarinic antagonist monotherapy to triple therapy in patients with COPD: a retrospective THIN-database analysis.
      ,
      • Jacobsen R.
      • Ekholm O.
      • Rasmussen N.K.
      • Hansen E.H.
      • Frolich A.
      Socioeconomic variations in use of prescription Medicines for COPD: a register-based study.
      ,
      • Di Martino M.
      • Agabiti N.
      • Bauleo L.
      • et al.
      Use patterns of long-acting bronchodilators in routine COPD care: the OUTPUL study.
      ] showed a general under usage of bronchodilator therapy in comparison to inhaled steroids, ICS monotherapy is generally not recommended for COPD treatment, due to higher risk of side effects (such as pneumonia) and absence of effect on long-term decline of FEV1 values [
      • Vogelmeier C.F.
      • Criner G.J.
      • Martinez F.J.
      • et al.
      Global strategy for the diagnosis, management, and prevention of chronic obstructive lung disease 2017 report. GOLD executive summary.
      ]. However, our study shows that ICS monotherapy was inversely associated with the escalation to triple therapy. While some positive effects of ICS monotherapy on exacerbations’ frequency have been shown, it is possible that part of this effect is explained by misdiagnosis of COPD and consequently by a lower or reduced trend to clinical deterioration of these patients compared to severe COPD. Furthermore, a low rate of specialist referral might explain both the high proportion of patients prescribed with ICS as starting therapy and the lower incidence of triple therapy in this group. Furthermore, we were able to stratify according to COPD GOLD stages less than the 10% of patients in our study. This finding might be explained by an under-use of spirometry tests to objectify airflow limitation, as already reported by Nishi SP et al. [
      • Nishi S.P.
      • Wang Y.
      • Kuo Y.F.
      • Goodwin J.S.
      • Sharma G.
      Spirometry use among older adults with chronic obstructive pulmonary disease: 1999-2008.
      ]. Anyhow, it is also plausible that spirometry data were available to GPs, but not recorded in the system. Both conditions might, anyway, have detrimental impact on the diagnosis, therapy and follow-up of COPD [
      • Anzueto A.
      • Miravitlles M.
      Considerations for the correct diagnosis of COPD and its management with bronchodilators.
      ,
      • Tinkelman D.G.
      • Price D.B.
      • Nordyke R.J.
      • Halbert R.J.
      Misdiagnosis of COPD and asthma in primary care patients 40 years of age and over.
      ,
      • Walters J.A.
      • Walters E.H.
      • Nelson M.
      • et al.
      Factors associated with misdiagnosis of COPD in primary care.
      ].
      A few limitations of the present study are worthy to be mentioned. First, because of the nature of the database, we were able to include among the potential predictors only a limited number of variables. We cannot exclude residual confounding derived from uncollected data. However, the measures we used are those commonly available in primary care databases, which practitioners base their decisions upon. Second, the high number of missing data for some conditions, such as obesity and smoking, can modify the strength of association. However, using as the reference category a group of patients with similar characteristics, this bias is minimized. As we followed up a homogenous cohort of patients receiving COPD treatment, the distribution of patients with missing values is unlikely to be different between groups. In this setting, using “missing” categories in the analysis is likely to be a reasonable approach [
      • Delaney J.A.
      • Moodie E.E.
      • Suissa S.
      Validating the effects of drug treatment on blood pressure in the General Practice Research Database.
      ]. Third, despite the availability of the 2011 update of the GOLD guidelines, in the present study COPD severity was staged according to the 2007 version, potentially limiting the generalizability of our results to actual clinical practices. However, being most of the collected data antecedent 2011, and considering that the observed prescription patterns may have been influenced by the COPD staging system in use at that time, we believe reasonable to use the old classification. To note, the information on number and the severity of the exacerbations – which are included in the 2011 GOLD recommendations for staging COPD severity – are included among the potential predictors considered in the study.
      In conclusion, our study showed that the incidence of triple inhaled therapy in primary care COPD patients in Italy is similar to the one found in other countries. The fact that patients with more severe COPD and more frequent or severe exacerbations are those at increased risk of escalation towards full inhaled therapy, suggests that GOLD recommendations reflect Italian GPs’ criteria for the escalation to triple therapy, and supports the rationale of an earlier triple inhaled therapy prescription, already as first approach in those with history of hospitalizations for respiratory infections. Anyhow, the high prescription of ICSs as starting therapy, might suggest a low awareness of clinical management of COPD in its early stages. Knowing real-life determinants of escalation to triple COPD therapy might help clinicians to personalize treatments of respiratory patients and international societies to develop guidelines suitable for real world challenges.

      Funding

      This study was funded by Chiesi Farmaceutici S.p.A., which approved the study design but played no role in development of the research and manuscript, including data analysis, results interpretation and writing.

      Disclosure statement

      FL provided consultancies in protocol preparation for epidemiological studies and data analyses for Chiesi, Novartis and GSK; CC provided clinical consultancies for Chiesi, Novartis and GSK, AP and MZ are Chiesi's employees. DLV, AZ, EB, AR and GO have no conflict of interests to disclose.

      Appendix A. Supplementary data

      The following are the Supplementary data to this article:Supplementary material.

      References

        • Adeloye D.
        • Chua S.
        • Lee C.
        • Basquill C.
        • Papana A.
        • Theodoratou E.
        • Nair H.
        • Gasevic D.
        • Sridhar D.
        • Campbell H.
        • Chan K.Y.
        • Sheikh A.
        • Rudan I.
        • Global Health Epidemiology Reference Group (GHERG)
        Global and regional estimates of COPD prevalence: systematic review and meta-analysis.
        J Glob Health. 2015 Dec; 5020415
        • Murray C.J.
        • Lopez A.D.
        Alternative projections of mortality and disability by cause 1990-2020: global burden of disease study.
        Lancet. 1997; 349: 1498-1504
        • Calverley P.M.
        • Anderson J.A.
        • Celli B.
        • et al.
        Salmeterol and fluticasone propionate and survival in chronic obstructive pulmonary disease.
        N. Engl. J. Med. 2007; 356: 775-789
        • Tashkin D.P.
        • Celli B.
        • Senn S.
        • et al.
        A 4-year trial of tiotropium in chronic obstructive pulmonary disease.
        N. Engl. J. Med. 2008; 359: 1543-1554
        • Mahler D.A.
        • Donohue J.F.
        • Barbee R.A.
        • et al.
        Efficacy of salmeterol xinafoate in the treatment of COPD.
        Chest. 1999; 115: 957-965
        • Gold P.M.
        The 2007 GOLD Guidelines: a comprehensive care framework.
        Respir. Care. 2009; 54: 1040-1049
        • Rodriguez-Roisin R.
        • Rabe K.F.
        • Vestbo J.
        • Vogelmeier C.
        • Agusti A.
        Global initiative for chronic obstructive lung disease (GOLD) 20th anniversary: a brief history of time.
        Eur. Respir. J. 2017; 50
        • Lipson D.A.
        • Barnhart F.
        • Brealey N.
        • et al.
        Once-Daily single-inhaler triple versus dual therapy in patients with COPD.
        N. Engl. J. Med. 2018; 378: 1671-1680
        • Singh D.
        • Papi A.
        • Corradi M.
        • et al.
        Single inhaler triple therapy versus inhaled corticosteroid plus long-acting beta2-agonist therapy for chronic obstructive pulmonary disease (TRILOGY): a double-blind, parallel group, randomised controlled trial.
        Lancet. 2016; 388: 963-973
        • Price D.
        • West D.
        • Brusselle G.
        • et al.
        Management of COPD in the UK primary-care setting: an analysis of real-life prescribing patterns.
        Int. J. Chronic Obstr. Pulm. Dis. 2014; 9: 889-904
        • Ernst P.
        • Saad N.
        • Suissa S.
        Inhaled corticosteroids in COPD: the clinical evidence.
        Eur. Respir. J. 2015; 45: 525-537
        • DiMarco F.
        • Santus P.
        • Terraneo S.
        • et al.
        Characteristics of newly diagnosed COPD patients treated with triple inhaled therapy by general practitioners: a real world Italian study.
        NPJ Prim Care Respir Med. 2017; 27: 51
        • Vetrano D.L.
        • Bianchini E.
        • Onder G.
        • et al.
        Poor adherence to chronic obstructive pulmonary disease medications in primary care: role of age, disease burden and polypharmacy.
        Geriatr. Gerontol. Int. 2017; 17: 2500-2506
        • Vogelmeier C.F.
        • Criner G.J.
        • Martinez F.J.
        • et al.
        Global strategy for the diagnosis, management, and prevention of chronic obstructive lung disease 2017 report. GOLD executive summary.
        Am. J. Respir. Crit. Care Med. 2017; 195: 557-582
        • Bloom C.I.
        • Elkin S.L.
        • Quint J.K.
        Changes in COPD inhaler prescriptions in the United Kingdom, 2000 to 2016.
        Int. J. Chronic Obstr. Pulm. Dis. 2019; 14 (Epub 2019/02/19. PubMed PMID: 30774325; PubMed Central PMCID: PMCPMC6349076): 279-287https://doi.org/10.2147/copd.s190086
        • Hurst J.R.
        • Dilleen M.
        • Morris K.
        • Hills S.
        • Emir B.
        • Jones R.
        Factors influencing treatment escalation from long-acting muscarinic antagonist monotherapy to triple therapy in patients with COPD: a retrospective THIN-database analysis.
        Int. J. Chronic Obstr. Pulm. Dis. 2018; 13: 781-792
        • Brusselle G.
        • Price D.
        • Gruffydd-Jones K.
        • et al.
        The inevitable drift to triple therapy in COPD: an analysis of prescribing pathways in the UK.
        Int. J. Chronic Obstr. Pulm. Dis. 2015; 10: 2207-2217
        • Decramer M.L.
        • Chapman K.R.
        • Dahl R.
        • et al.
        Once-daily indacaterol versus tiotropium for patients with severe chronic obstructive pulmonary disease (INVIGORATE): a randomised, blinded, parallel-group study.
        Lancet Respir Med. 2013; 1: 524-533
        • Nannini L.J.
        • Poole P.
        • Milan S.J.
        • Holmes R.
        • Normansell R.
        Combined corticosteroid and long-acting beta(2)-agonist in one inhaler versus placebo for chronic obstructive pulmonary disease.
        Cochrane Database Syst. Rev. 2013; Cd003794
        • Hawkins N.M.
        • Virani S.
        • Ceconi C.
        Heart failure and chronic obstructive pulmonary disease: the challenges facing physicians and health services.
        Eur. Heart J. 2013; 34: 2795-2803
        • Simeone J.C.
        • Luthra R.
        • Kaila S.
        • et al.
        Initiation of triple therapy maintenance treatment among patients with COPD in the US.
        Int. J. Chronic Obstr. Pulm. Dis. 2017; 12: 73-83
        • Mapel D.
        • Laliberte F.
        • Roberts M.H.
        • et al.
        A retrospective study to assess clinical characteristics and time to initiation of open-triple therapy among patients with chronic obstructive pulmonary disease, newly established on long-acting mono- or combination therapy.
        Int. J. Chronic Obstr. Pulm. Dis. 2017; 12: 1825-1836
        • Sundh J.
        • Aberg J.
        • Hasselgren M.
        • et al.
        Factors influencing pharmacological treatment in COPD: a comparison of 2005 and 2014.
        Eur Clin Respir J. 2017; 4: 1409060
        • Souliotis K.
        • Kani C.
        • Papageorgiou M.
        • Lionis D.
        • Gourgoulianis K.
        Using big data to assess prescribing patterns in Greece: the case of chronic obstructive pulmonary disease.
        PLoS One. 2016; 11e0154960
        • Carrasco-Garrido P.
        • de Miguel-Diez J.
        • Rejas-Gutierrez J.
        • et al.
        Characteristics of chronic obstructive pulmonary disease in Spain from a gender perspective.
        BMC Pulm. Med. 2009; 9: 2
        • Tsiligianni I.
        • Rodriguez M.R.
        • Lisspers K.
        • LeeTan T.
        • Infantino A.
        Call to action: improving primary care for women with COPD.
        NPJ Prim Care Respir Med. 2017; 27: 11
        • Jacobsen R.
        • Ekholm O.
        • Rasmussen N.K.
        • Hansen E.H.
        • Frolich A.
        Socioeconomic variations in use of prescription Medicines for COPD: a register-based study.
        Respir. Care. 2016; 61: 943-949
        • Di Martino M.
        • Agabiti N.
        • Bauleo L.
        • et al.
        Use patterns of long-acting bronchodilators in routine COPD care: the OUTPUL study.
        COPD. 2014; 11: 414-423
        • Nishi S.P.
        • Wang Y.
        • Kuo Y.F.
        • Goodwin J.S.
        • Sharma G.
        Spirometry use among older adults with chronic obstructive pulmonary disease: 1999-2008.
        Ann Am Thorac Soc. 2013; 10: 565-573
        • Anzueto A.
        • Miravitlles M.
        Considerations for the correct diagnosis of COPD and its management with bronchodilators.
        Chest. 2018; 154: 242-248
        • Tinkelman D.G.
        • Price D.B.
        • Nordyke R.J.
        • Halbert R.J.
        Misdiagnosis of COPD and asthma in primary care patients 40 years of age and over.
        J. Asthma. 2006; 43: 75-80
        • Walters J.A.
        • Walters E.H.
        • Nelson M.
        • et al.
        Factors associated with misdiagnosis of COPD in primary care.
        Prim. Care Respir. J. 2011; 20: 396-402
        • Delaney J.A.
        • Moodie E.E.
        • Suissa S.
        Validating the effects of drug treatment on blood pressure in the General Practice Research Database.
        Pharmacoepidemiol. Drug Saf. 2008; 17: 535-545