Clinical characteristics governing treatment adjustment in COPD patients: results from the Swiss COPD cohort study

DOI: https://doi.org/https://doi.org/10.57187/smw.2023.40114

Introduction

Chronic obstructive pulmonary disease (COPD) is a widespread chronic disease and places a major burden on patients and healthcare systems. The disease is an increasing cause of morbidity and mortality, in contrast to other major chronic diseases [2]. With a global prevalence of approximately 11.7%, COPD is considered the fourth leading cause of death worldwide [3]. COPD is also associated with a significant economic burden by producing an annual cost of nearly 50 billion Euros in the EU [2]. COPD is a major problem globally because of increasing tobacco consumption and pollution along with ageing of the world’s population [4].

In response to the burden of COPD, the Global Initiative for Chronic Obstructive Lung Disease (GOLD) created guidelines to help optimise the treatment of COPD, increase awareness of the disease and decrease morbidity and mortality. The GOLD guidelines provide advice on how to optimise COPD prevention, diagnosis and treatment [5].

A COPD diagnosis is made according to the patient’s established medical history. In this longitudinal cohort study, the physician noted each patient’s medical history and conducted a medical examination. All patients were asked to complete the COPD Assessment Test (CAT) questionnaire. The COPD Assessment Test score was calculated by summing the points for each variable. This resulted in a CAT score of 0–40, classified into four groups ranging from low to very high, depending on the level of impact of the disease on the person’s health. Patients in this cohort study were typically categorised depending on the severity of airflow limitation, the number of exacerbations within the last year and the current impact of symptoms, as assessed by the CAT score. Depending on the severity of airway obstruction, patients were assigned GOLD 1, 2, 3 or 4. This categorisation would then aid the physician’s diagnosis and treatment of the patient in question. In a revision of the GOLD guidelines in 2017, a refinement of the ABCD tool was proposed, separating spirometric grades from the ABCD groups. In this refined assessment scheme, patients should undergo spirometry to determine the severity of airflow limitation and then undergo an assessment of either dyspnea using the modified Medical Research Council dyspnea scale (mMRC) or symptoms using COPD Assessment Test. The history of exacerbations should also be recorded.

Pharmacological treatments according to GOLD 2017 recommendations were defined as follows:

• Group A: short-acting beta-agonist (SABA) or short-acting muscarinic antagonist (SAMA)
• Groups B and C: long-acting beta-agonist (LABA), long-acting muscarinic antagonist (LAMA) or LAMA/LABA combination
• Group D: any treatment regimen including LAMA/LABA combination

While guidelines are essential for the standardisation of disease, clinical practice often differs from proposed treatment recommendations [6]. Therefore, this study sought to gain insight into the factors governing a physician’s decision to adjust therapy for COPD patients.

The under- and over-diagnosis of COPD in the primary healthcare sector likely contribute to the increase in morbidity and mortality associated with this disease. It remains unclear how much the integration of GOLD guidelines affects the clinical outcomes of COPD patients. Although patient-reported outcome questionnaires like the CAT and the mMRC are necessary to assess the severity of the disease and to categorise patients correctly so that they receive the recommended treatments, it seems that a majority of physicians in real practice settings do not use these questionnaires [3]. This might lead to physicians underestimating symptoms and thus not adjusting therapy as might be necessary. Several cases of misdiagnosis have been noted in Switzerland, including in patients who have been diagnosed with COPD but have not fulfilled the GOLD criteria [2]. Patients with COPD are typically not referred to specialists as often as are patients with other chronic conditions, such as diabetes or chronic heart disease. Therefore, to accurately assess the prevalence of COPD, it is necessary to gain better insight into the management of COPD therapy, which typically falls to the primary care sector.

Hypothesis and main goals

The focus of this study was gaining insight into the management of COPD therapy within the primary care sector. More specifically, the goal of the study was to investigate factors associated with changes in COPD therapy.

Materials and methods

This project used data from the Swiss COPD Cohort Study, which is an ongoing, multicentric, population-based study with the aim of assessing COPD management within the primary care setting. Participants have been recruited since 2007. The present study focused on patients recruited from 2015–2019. General practitioners from Switzerland were invited to join the study. In total, 24 general practitioners participated in the Swiss COPD Cohort Study. The ethics committees of all cantons involved permitted the study (EK Nr. 170/06). The 24 general practitioners were asked to invite all patients with mild to severe COPD (GOLD stages 1–4) to participate in the study. All patients included in the study provided informed consent. There was an initial study examination (baseline visit), followed by re-examinations every 6 months. The inclusion criteria were as follows:

• FEV1/FVC <70% after inhalation of a bronchodilator
• age >40 years
• smoker or ex-smoker of at least 20 pack-years
• informed consent

During the baseline visit, general information, like age, sex, weight, height, year of first diagnosis of COPD, smoking status, number of attempts to stop smoking and pack-years, was assessed. General practitioners used a standardised questionnaire to document symptoms by assessing the CAT score and mMRC, comorbidities, exacerbations, spirometry and treatment. The following comorbidities were evaluated using the questionnaire: asthma, cardiovascular diseases (coronary heart disease, heart failure, peripheral artery disease, hypertension and cerebrovascular insult), diabetes and malignant diseases.

General practitioners then conducted a spirometric assessment according to European Respiratory Society (ERS) / American Thoracic Society (ATS) technical standards on interpretative strategies for routine lung function tests [7]. They documented the forced expiratory volume in 1 second (FEV1) in litres; FEV1 percentage of the reference value and/or lower limit of normal; forced vital capacity (FVC) in litres; FVC percentage of the reference value and/or lower limit of normal; and FEV1/FVC (%) (Tiffeneau index). During follow-up visits, patients’ symptoms were evaluated by assessing the CAT score and mMRC [4, 5] and by asking about medication, non-pharmacological treatments, hospitalisation and exacerbations. Undesired changes were defined as a worsening of symptoms that may have required a change in therapy.

The patients’ data were anonymised and were entered upon collection into a central electronic database (RDE Light) by the study team. Patients with at least one additional visit within one year of the baseline visit were included. Patients with a change in therapy were signalled to the study team, in line with the study’s objective. Our primary outcome was a change in therapy, which was defined as a change in either medication (± medication) or non-medical treatment, such as pulmonary rehabilitation. A change in medication was only counted if there was a change of substance class; i.e., a change within one substance class was not regarded as a change in therapy. Adjustments to therapy were categorised into three groups: step up in therapy, step down, and step up and step down at the same time. An example of a step up and step down at the same time is the new prescription of a long-acting muscarinic antagonist while treatment with a short-acting beta-agonist was stopped. For those patients for whom a change in treatment was detected, we sought to establish factors associated with the treatment adjustment.

Statistical analysis

Data were summarised using descriptive statistics (mean ± standard deviation for continuous data, absolute and relative frequencies for categorical data). The effects of disease characteristics on treatment change were analysed using separate univariable and multivariable logistic mixed models. These models included the following as independent variables: the disease characteristic (as a fixed effect) and the patient (as a random effect). We excluded mMRC from the multivariable regression because of its close association with GOLD groups. The effect of the disease characteristic on the change in treatment was calculated as an odds ratio (OR) with a 95% confidence interval (CI). The hypothesis of no effect was evaluated using the Wald test. P-values were interpreted in a descriptive sense.

Data analysis was conducted according to the standard operating procedures of the Clinical Trials Unit, University Medical Centre Freiburg. All statistical analyses were performed using SAS version 9.4 (SAS Institute, Inc.). For descriptive analyses, the procedures FREQ and MEAN were used. For the generalised mixed models, the procedure GLIMMIX was used. Analytical code is available from the authors upon reasonable request.

Ethical approval and consent to participate

We received ethical approval for this study by the local ethical committee (Ethikkommission Nordwest- und Zentralschweiz [EKNZ], formerly Ethikkommission beider Basel [EKBB]) in 2017 (EK Nr. 170/06)and subsequently by ethical committees of all other participating Swiss cantons. ClinicalTrials.gov Identifier: NCT02065921.

All participants provided written consent to participate in the study.

Results

During the study period, we recruited 260 patients, 253 of whom had received a spirometry diagnosis of COPD. A total of 48 patients were excluded from the study because they were missing an additional visit after the baseline visit or their visits were more than 1 year apart. A further 10 patients were excluded because of missing information on medication (figure 1).

Figure 1Study population. Summary of changes in treatment.

In total, patients made 690 visits. There were 500 visits that did not result in a change in therapy and 190 visits that resulted in a change in drug therapy (table 2). The table shows the odds ratios for predictors of change in therapy. The majority of therapy adjustments were to medication. Only 9.2% of the patients received pulmonary rehabilitation at any point within the study period. The changes in therapy were categorised into three groups, namely a step up in treatment, a step down, and a step up and a step down at the same time. During 72% of the visits, no treatment adjustments were made. In those cases, medication was not changed, and no pulmonary rehabilitation was prescribed. During 28% (190) of the visits, general practitioners modified medication or prescribed pulmonary rehabilitation. For 94/190 visits (49.5%) with a change in treatment, there was a step up in therapy, whereas in 62/190 cases (32.6%) there was a step down in therapy. In the remaining 34/190 cases (17.9%), there were both a step up and a step down in medication during one visit. Furthermore, we observed that 32.6% of the overall population received a clinical diagnosis of COPD that could not be confirmed by spirometry. 

Therapy increase

In the multivariable analysis, we found that the severity of disease was linked to the probability of experiencing a therapy increase: patients in GOLD Group C were more likely to experience an increase in therapy compared to patients in GOLD Group A (OR 3.34 [95% CI: 1.02–11.55; p = 0.013]). Also, compared to patients with mild obstruction, patients with severe (OR 4.24 [95% CI: 1.88–9.56]) and very severe (OR 5.48 [95% CI: 1.31–22.96]) obstruction were more likely to experience a therapy increase (p <0.0001). Comorbidities were negatively associated with increases in therapy (OR 0.42 [95% CI: 0.24–0.73; p = 0.002]) (table 3).

Therapy decrease

GOLD Group C was also associated with a therapy decrease (OR 9.23 [95% CI: 2.31–36.85; p = 0.0045]). The severity of airflow limitation was not associated with steps down in medication. Exacerbation was positively associated with a decrease in treatment (OR 2.66 [95% CI: 1.01–6.97; p = 0.0471]). A higher COPD Assessment Test score was mildly associated with a step down in medication (OR 1.07 [95% CI: 1.01–1.14; p = 0.014]). Comorbidities were not associated with a step down in therapy. Neither cough nor sputum showed an association with therapy decrease (table 4).

Therapy increase and decrease at the same time

Visits with a step up and step down at the same time were strongly associated with both very severe airflow limitation (OR 589 [95% CI: 2.72 – >999; p = 0.1092]) and exacerbation (OR 8.93 [95% CI: 1.17–68.28; p = 0.0349]). Neither GOLD group, CAT score nor the presence of cough and/or sputum was associated with both a therapy increase and decrease during one visit (table 5).

Discussion

This cohort study has provided deeper insight into how patients with COPD are treated in primary care settings. COPD Group C and severe to very severe airflow limitation were both associated with an increase in COPD treatment. Patients with comorbidities were more likely to have received no treatment changes. Past COPD exacerbations were not associated with therapy increases.

The fact that GOLD Group C patients were more likely to see a step up in treatment compared to GOLD A patients seemed consistent with expectations for clinical practice. This is because patients in GOLD Group C are affected more by the disease, as they experience more exacerbations, although they do not have more symptoms between exacerbations than do GOLD Group A patients. A therapy decrease was associated with GOLD Group C and the presence of exacerbations. This result was rather unexpected, as a therapy increase is expected in the case of more severe disease. That is, GOLD Group A or B is expected to be associated with a step down in therapy. There were no concrete associations between GOLD Group D and a change in therapy, likely due to the small number of patients in that group. Patient Group C was the smallest of our study. The size of GOLD Groups A–D can differ depending on the questionnaire used. COPD Assessment Test and mMRC correlate with each other; however, the inter-rater agreement between CAT and mMRC is quite good [8]. That might be a reason why patient Group C was associated with a treatment increase as well as a treatment decrease in our study. The newest COPD GOLD guidelines propose to merge groups C and D into a single group E [9]. This suggests that frequent exacerbators with few symptoms (Group C) should receive the same treatment as frequent exacerbators with high symptom load (Group D). Together with our finding that Group C was associated with all different types of therapy changes, this may imply that Group C is very difficult to treat satisfactorily. The newest GOLD suggestions may help in treating this particular patient group.

Our results showed that exacerbation was not strictly associated with an increase in therapy provided by general practitioners, which was unexpected. Other studies have reported positive associations [10, 11]. Campos et al. considered the prescribed treatment during a routine follow-up visit for COPD treatment. The results indicated that the majority of doctors did not change the patient’s treatment regimen. Among those whose therapeutic regimen changed, disease exacerbation was the main driver of increased treatment [11]. Although mMRC and the patient’s CAT score serve as indicators for assessment and (change in) treatment, our study did not find them to be associated with changes in therapy. The study noted that the presence of comorbidities made treatment changes less likely. This is highly interesting and relevant, as 71.3% of patients had comorbidities. We had expected the comorbidities to lead to therapy adjustments, as comorbidities potentially cause worsened COPD-related symptoms, like dyspnea, and more suffering due to the disease. Overington et al. explain such circumstances with presumed barriers to implementing guidelines, including factors such as lack of familiarity amongst clinicians with guidelines and inadequate implementation programs [12]. In patients with multimorbidity, an increase in dyspnea, and therefore in COPD Assessment Test or mMRC, is not necessarily caused by COPD itself. It is possible that physicians judged the increase in symptoms as not being related to lung disease [13].

Patient-reported questionnaires like the COPD Assessment Test and mMRC are necessary to assess the severity of the disease and to group patients correctly so that they receive the recommended treatment. However, a majority of physicians in real practice settings do not use these questionnaires [14]. That might lead to physicians underestimating symptoms and thus not adjusting the therapy, as they perhaps should be doing. Studies such as Wang et al. have confirmed that that the prevalence of excess polypharmacy increases with age, thereby increasing the risk of various clinical outcomes. This study assumed that general practitioners hesitate to adjust treatment in COPD patients due to these potential side effects, as suggested by studies in the field [14]. Other suggested reasons for this hesitation were a potential lack of time or even patient hesitancy in changing medication. Therefore, this study suggested further educational efforts to disseminate knowledge of GOLD guidelines to improve the implementation of COPD management recommendations.

Our data showed that a number of patients were diagnosed with COPD without fulfilling spirometric criteria and even had treatment adjusted (step up or step down), only partially in line with GOLD suggestions. This somewhat ad hoc implementation of GOLD guidelines for COPD management in a primary care setting is the focus of Sehl et al.’s literature review of adherence to COPD management guidelines in general practice [15]. Sehl et al. showed variability in which parts of the guidelines general practitioners adhered to. One of the most commonly cited gaps in adherence was the omission of spirometry. Spirometry is essential and required to confirm a COPD diagnosis [10]. The omission is therefore contrary to the GOLD guidelines [16].

The data presented in our study tend to agree with similar studies that have investigated the working integration of GOLD guidelines into COPD management in the primary care sector. Studies such as that of Marmy et al. reached similar conclusions to our study, for example by reporting that up to 53% of COPD patients in their study group were not consistently treated according to GOLD 2017 recommendations [3]. Jochmann et al. also highlighted the manner in which COPD treatment in primary care tends to differ from proposed GOLD treatment options [17]. This study raised the issue of general practitioners not using the COPD Assessment Test questionnaire, which is recommended by the GOLD guidelines for an adequate assessment of patients. Similarly, Jochmann et al. highlighted a lack of the spirometry that should be performed on a routine basis [17]. A further study by Urwyler et al. underlined the importance of fulfilling spirometric criteria. Since all spirometry tests are administered by the respective primary physician, we were unable to account for this bias. The results of this study raise the question of the extent of GOLD guideline knowledge and suggest the need for more detailed knowledge dissemination among general practitioners [18]. Grewe et al. found that the prescription of medicine with treatment adjustment was not as would be considered suitable for the patient’s symptoms and was only partially compliant with GOLD guidelines. Patients with more severe COPD were more likely to be treated according to the guidelines, which resulted in a low rate of adherence therapy (59.1%), in line with the results of other studies [19]. This study noted the general lack of concrete data on COPD management with general practitioner surgeries, which would provide valuable insight on the subject.

Conclusion

The authors note the following weaknesses of the present study. This project is likely to have drawn some inaccurate conclusions due to some inconclusive data. Some patients had to be excluded due to missing/inconclusive data. Furthermore, patients who did not attend a follow-up visit were excluded from the analysis, which may have resulted in bias. The patients that were recruited by the general practitioner participated in the study voluntarily, which could have resulted in a selection bias leading to a sample unrepresentative of the population. This study did not ask the physicians involved for the motivations behind their actions, which may be an interesting area for further study.

Data availability statement

The datasets used and analysed in this study are available from the corresponding authors upon reasonable request.

Acknowledgements

We thank all general practitioners and pulmonologists who took part in the study. Without their extra efforts, this research would not have been possible.

_Toc472330565Author contributions: LK: Conception of study and design, interpretation of data, acquisition and analysis of data and contributing to the manuscript. ZP: Writing consequent drafts of the manuscript. NAH, POB, PNC: acquisition of data, support for the participants. TG, LJZ, MK, SM, DM, MT, RT, CvG: Contributions to the conception of the study, analysis and interpretation of the data, contributions to the manuscripts. JDL: Responsible principle investigator, supervision of acquisition of data, contributions to the manuscript.

References

1. Global Initiative for Chronic Obstructive Lung Disease [Internet]. 2019. Available from: www.goldcopd.org 

2. Román-Rodríguez M, Pardo MG, López LG, Ruiz AU, van Boven J. Enhancing the use of Asthma and COPD Assessment Tools in Balearic Primary Care (ACATIB): a region-wide cluster-controlled implementation trial. NPJ Prim Care Respir Med. 2016 Mar;26(1):16003. 10.1038/npjpcrm.2016.3 

3. Marmy JL, Diedrich JP, Cadus C, Grendelmeier P, Tschacher A, Dieterle T, et al. Adherence to GOLD Recommendations among Swiss Pulmonologists and General Practitioners. COPD. 2021 Feb;18(1):9–15. 10.1080/15412555.2020.1859469 

4. Jones PW, Harding G, Berry P, Wiklund I, Chen WH, Kline Leidy N. Development and first validation of the COPD Assessment Test. Eur Respir J. 2009 Sep;34(3):648–54. 10.1183/09031936.00102509 

5. Smid DE, Gonik M, Miravitlles M, Casanova C, Cosio BG, de Lucas-Ramos P, et al. Redefining cut-points for high symptom burden of the GOLD classification in 18577 patients with chronic obstructive pulmonary disease. 

6. Lõpez-Campos JL, Tan W, Soriano JB. Global burden of COPD. Vol. 21, Respirology. Blackwell Publishing; 2016. p. 14–23. 10.1111/resp.12660 

7. Stanojevic S, Kaminsky DA, Miller MR, Thompson B, Aliverti A, Barjaktarevic I, et al. ERS/ATS technical standard on interpretive strategies for routine lung function tests. Eur Respir J. 2022 Jul;60(1):2101499. 10.1183/13993003.01499-2021 

8. Zogg S, Dürr S, Miedinger D, Steveling EH, Maier S, Leuppi JD. Differences in classification of COPD patients into risk groups A-D: a cross-sectional study. BMC Res Notes. 2014 Aug;7(1):562. 10.1186/1756-0500-7-562 

9. Global Initiative for Chronic Obstructive Lung Disease Inc. Global Initiative for Chronic Obstructive Lung Disease: Global Strategy for Diagnosis, Management, and prevention of Obstructive Pulmonary Disease. 

10. Yawn BP, Mintz ML, Doherty DE. Gold in practice: chronic obstructive pulmonary disease treatment and management in the primary care setting. Int J Chron Obstruct Pulmon Dis. 2021 Feb;16:289–99. 10.2147/COPD.S222664 

11. López-Campos JL, Abad Arranz M, Calero Acuña C, Romero Valero F, Ayerbe García R, Hidalgo Molina A, et al. Determinants for changing the treatment of COPD: a regression analysis from a clinical audit. Int J Chron Obstruct Pulmon Dis. 2016 Jun;11(1):1171–8. 10.2147/COPD.S103614 

12. Overington JD, Huang YC, Abramson MJ, Brown JL, Goddard JR, Bowman RV, et al. Implementing clinical guidelines for chronic obstructive pulmonary disease: barriers and solutions. J Thorac Dis. 2014 Nov;6(11):1586–96. 

13. Reilev M, Pottegård A, Lykkegaard J, Søndergaard J, Ingebrigtsen TS, Hallas J. Increased risk of major adverse cardiac events following the onset of acute exacerbations of COPD. Respirology. 2019 Dec;24(12):1183–90. 10.1111/resp.13620 

14. Wang R, Chen L, Fan L, Gao D, Liang Z, He J, et al. Incidence and effects of polypharmacy on clinical outcome among patients aged 80+: A five-year follow-up study. PLoS One. 2015 Nov;10(11):e0142123. 10.1371/journal.pone.0142123 

15. Sehl J, O’Doherty J, O’Connor R, O’Sullivan B, O’Regan A. Adherence to COPD management guidelines in general practice? A review of the literature. Ir J Med Sci. 2018 May;187(2):403–7. 10.1007/s11845-017-1651-7 

16. Gruffydd-Jones K, Brusselle G, Jones R, Miravitlles M, Baldwin M, Stewart R, et al. Changes in initial COPD treatment choice over time and factors influencing prescribing decisions in UK primary care: In UK primary care: A real-world, retrospective, observational. Vol. 26, npj Primary Care Respiratory Medicine. Nature Publishing Group; 2016. 

17. Jochmann A, Neubauer F, Miedinger D, Schafroth S, Tamm M, Leuppi JD. General practitioner’s adherence to the COPD GOLD guidelines: baseline data of the Swiss COPD Cohort Study. Swiss Med Wkly. 2010 Aug;140(AUGUST):10.4414/smw.2010.13053. 10.4414/smw.2010.13053 

18. Urwyler P, Abu Hussein N, Bridevaux PO, Chhajed PN, Geiser T, Grendelmeier P, et al. Predictive factors for exacerbation and re-exacerbation in chronic obstructive pulmonary disease: an extension of the Cox model to analyze data from the Swiss COPD cohort. Multidiscip Respir Med. 2019 Feb;14(1):7. 10.1186/s40248-019-0168-5 

19. Grewe FA, Sievi NA, Bradicich M, Roeder M, Brack T, Brutsche MH, et al. Compliance of Pharmacotherapy with GOLD Guidelines: A Longitudinal Study in Patients with COPD. Int J Chron Obstruct Pulmon Dis. 2020 Mar;15:627–35. 10.2147/COPD.S240444