Abstract
The use of beta-blockers in patients with chronic obstructive pulmonary disease (COPD) has received much attention. Several observational studies reported important reductions in mortality and exacerbations with these drugs, but the extent of bias in these studies is unclear. Nevertheless, the large ongoing randomized trial (βLOCK-COPD) was initiated specifically to evaluate these effects. We searched the literature to identify all observational studies investigating the effectiveness of beta-blockers in COPD patients on major outcomes, including death and COPD exacerbation. We identified 18 observational studies, with 10 studies affected by confounding bias and six by immortal time bias, while two addressed these biases. Reductions in all-cause mortality with beta-blocker use were observed among the studies with confounding bias (pooled rate ratio 0.72; 95% CI 0.59–0.88) and those with immortal time bias (pooled rate ratio 0.64; 95% CI 0.53–0.77). A large five-database study that addressed these two biases reported hazard ratios of 0.90 (95% CI: 0.78–1.02) for death and 0.54 (95% CI: 0.47–0.61) for COPD hospitalization. However, this latter estimate was the same as for the first 30 days after treatment initiation, thus indicating that important residual confounding cannot be ruled out. Observational studies, important to provide evidence from real-world data on medication effects, are unsupportive for beta-blockers in COPD. Even if immortal time bias is properly avoided, confounding bias cannot be fully controlled due to their relative contraindication in COPD. In the case of beta-blockers, randomized trials such as βLOCK-COPD are necessary to eliminate the uncertainty from residual confounding bias.
Keywords:
Introduction
The use of beta-blockers in patients with chronic obstructive pulmonary disease (COPD) has posed a challenge in view of the risk of bronchospasm and neutralization of the effectiveness of β2-agonists (Citation1). This is important since patients with COPD have a high prevalence of cardiovascular comorbidity (Citation2). Many of these patients, in the absence of COPD, would normally receive and benefit from beta-blockers, which have been shown to reduce mortality in patients with heart failure and other cardiovascular conditions (Citation3, Citation4). This question of beta-blocker use, as well as their potential adverse and beneficial effects in COPD, has been receiving considerable attention (Citation5). Recent re-analyses of the TORNADO and SUMMIT trials present results that provide some reassurance as to the safety of concomitant use beta-blockers along with long-acting bronchodilators in COPD (Citation6, Citation7).
Beyond safety, several observational studies reported important reductions in mortality and exacerbations with beta-blocker use. A meta-analysis of nine studies reported a relative risk of COPD-related mortality of 0.69 (95% CI: 0.62–0.78) with beta-blocker use versus nonuse (Citation8). A second meta-analysis of 15 studies reported a relative risk of all-cause mortality and of COPD exacerbation of 0.72 (95% CI: 0.63–0.83) and 0.63 (95% CI: 0.57–0.71), respectively, with beta-blocker use versus nonuse (Citation9). As a result of this potential benefit, a large randomized trial (βLOCK-COPD) is currently underway to evaluate these effects (Citation10). It will enroll over 1000 patients with moderate to severe COPD who will be randomly allocated to the beta-blocker metoprolol or placebo and followed for a year for exacerbations, mortality and other outcomes.
However, while the authors of the meta-analyses of observational studies mentioned the potential for biases, this discussion was general and did not address specific biases for specific studies. In this paper, we review the observational studies of beta-blockers in COPD to assess the potential sources of bias.
Methods
We searched the literature for all publications of observational studies investigating the effectiveness of beta-blockers in COPD patients on two major COPD outcomes, namely COPD exacerbations and mortality. We searched PubMed for keywords “beta blockers” or “beta-blockers” and “COPD” with no restriction on publication date, and also used studies included in the meta-analyses. We screened the titles and abstracts, and carried forward potentially relevant publications that we reviewed and classified according to the potential biases induced by the study design, including immortal time bias and confounding.
Results
We identified 232 potential papers, of which we found that 212 were editorials or opinion pieces, studies of other drugs where beta-blockers were mentioned but not analyzed, studies where no results were presented for exacerbation or mortality, or studies of drug utilization. We thus reviewed in depth the relevant 20 studies that specifically addressed beta-blocker effectiveness in COPD on the outcomes of interest. We found that two studies, both based on the Taiwan databases, were insufficiently detailed in their methods to allow a proper evaluation, and were thus excluded (Citation11, Citation12). Of the remaining 18 studies, we found that many of these were subject to confounding bias or affected by immortal time bias.
Confounding bias
We found 10 of the studies that investigated the effects of beta-blockers in COPD to be subject to potential confounding bias () (Citation13–22). Several of the studies used data from existing randomized trials of drugs other than beta-blockers, but which included information on beta-blocker use, or registries of patients with cardiovascular diseases. Inherently, these data sources included extensive information on cardiovascular prognostic factors but little or no data on the severity of COPD. Thus, the opportunities to adjust for COPD severity when comparing users and nonusers of beta-blockers among the patients with COPD was greatly limited.
Table 1. Observational studies investigating the effects of beta-blockers in COPD on the risk of death or COPD exacerbation, affected by confounding bias.
An example of a study with residual confounding is the first study to investigate this question. It used the Cooperative Cardiovascular Project database of patients hospitalized with a principal diagnosis of acute myocardial infarction under Medicare in the US (Citation13). In the subgroup of patients with COPD, the adjusted relative risk (RR) of death from any cause at 2 years comparing users of beta-blockers with nonusers of beta-blockers at discharge was 0.60 (95% CI: 0.57–0.63). This estimate was adjusted for age, sex, ethnicity, co-morbidity, as well as an extensive number of cardiovascular factors, including several measures of physiologic condition at admission, the type of myocardial infarction, history of myocardial infarction, revascularization and heart failure. It also adjusted for cardiovascular treatment, including aspirin, angiotensin-converting enzyme inhibitors, calcium-channel blockers, heparin, thrombolytic agents, coronary angioplasty and bypass surgery. However, there were no COPD-specific prognostic factors included in adjusting the effect of beta-blockers among the COPD subgroup.
A subsequent reanalysis of this database, focusing specifically on the subgroup of patients with COPD or asthma (over 90% are COPD), attempted to account for COPD severity by using beta-agonists, oral steroids and COPD hospitalization as proxies (Citation14). Among patients not prescribed beta-agonists, the RR of death comparing users and nonusers of beta-blockers was 0.86 (95% CI: 0.73–1.00). Among those prescribed beta-agonists the RR was 0.88 (95% CI: 0.69–1.14) and among the more severe patients who received oral steroids or had been hospitalized for COPD or asthma in the previous year, the RR was 1.07 (95% CI: 0.75–1.52). Pooling across these three crude severity strata produces a RR of 0.89 (95% CI: 0.79–1.00), very different from the original estimate of 0.60 (95% CI: 0.57–0.63) which was not adjusted for COPD-specific prognostic factors and thus biased from confounding by COPD severity.
One of the main reasons for this confounding is that beta-blockers were prescribed to 37% of patients with COPD or asthma not on beta-agonists, 25% of those on beta-agonists, and 12% of the patients with severe COPD or asthma (Citation14). This pattern of use is inversely proportional to the expected increasing mortality with COPD severity. It is essential to account for this reluctance to prescribe beta-blockers with increasing severity of airways disease. Such confounding by contraindication of beta-blocker use in COPD biases the estimate of the effect of these drugs.
Some studies included more extensive adjustment for COPD-related confounders, yet still somewhat incomplete. For example, the study based on the COPDGene cohort had extensive data on respiratory morbidity at baseline, including the 6-minute walk distance, the St-George Respiratory Questionnaire (SGRQ) score and the Modified Medical Research Council (MMRC) dyspnea score (Citation20). In addition, medication history on all respiratory inhalers was available. The adjustment for confounding included, besides demographic and cardiovascular factors, the following COPD-specific factors: FEV1, %emphysema on CT and use of long-acting respiratory medications. However, the adjustment did not include prior exacerbations, home oxygen therapy and systemic steroids, all important prognostic factors in COPD and with different prevalence between the users and nonusers of beta-blockers. For example, 21 and 2.8% of the users of beta-blockers were on oxygen therapy and systemic steroids, respectively, compared with 29 and 6.1% for the nonusers of beta-blockers. These differences were not adjusted for (Citation20).
Immortal time bias
We found six studies affected by immortal time bias () (Citation23–28). In cohort studies, immortal time refers to a period of follow-up during which the outcome under study cannot occur (Citation29). Misclassifying or excluding this period in terms of exposure will introduce immortal time bias (Citation30).
Table 2. Observational studies investigating the effects of beta-blockers in COPD on the risk of death or exacerbation, affected by immortal time bias.
An example of immortal time bias from misclassification is the observational study by Rutten suggesting that the use of a beta-blocker is associated with a 32% reduction in all-cause mortality (Citation26). The study used electronic medical records from a network of 23 general practices in the Netherlands. The cohort included 2230 patients 45 years or older with a diagnosis of COPD between 1995 and 2005, followed until death, until they moved or the end of 2005. There were 665 patients who used beta-blockers during the cohort follow-up. Comparing time from diagnosis until death resulted in a 32% reduction in the rate of all-cause mortality (hazard ratio (HR) 0.68; 95% CI: 0.56–0.83) associated with beta-blocker use.
Immortal time bias was introduced in this study by considering the patients as exposed to beta-blockers from the day of diagnosis, even if they only filled their first prescription years later. The time between the first COPD diagnosis after 1995 and the first beta-blocker prescription during follow-up is immortal, as the patient must survive to receive this prescription. Moreover, this immortal period is unexposed to beta-blockers, so that the misclassification of this immortal unexposed period as exposed results in immortal time bias (Citation30). depicts this bias by comparing the survival times between typical users and nonusers, where the beta-blocker users will necessarily have a longer survival, artificially created by this immortal time in the beta-blocker users. This immortal time bias from exposure misclassification will result in an incorrectly “protective” effect of beta-blocker exposure.
Figure 1. Illustration of immortal time bias in cohort studies: The immortal time between the diagnosis of COPD (cohort entry) and the first beta-blocker prescription is misclassified as “exposed to beta-blocker” when, in fact ,the patient is unexposed.
Some of the studies identified beta-blocker use during hospitalization (Citation24, Citation25, Citation28). For example, Dransfield et al.’s (Citation24) study reported that inpatient use of beta-blockers by patients hospitalized for COPD is associated with a 61% reduction in mortality. Immortal time bias is introduced by defining exposure to beta-blockers by billings occurring at any time during the hospitalization (Citation31). Thus, the exposed patients necessarily had an initial period of their hospitalization with no exposure before they received their first beta-blocker. This period is immortal as the patient had to be alive to receive this drug. By defining exposure in this way, the immortal period conferred a survival advantage to the users of beta-blockers and an apparently longer survival. This likely explains in part the longer length of stay of 7.8 days for users of beta-blockers versus 5.3 days for nonusers (Citation24).
Quint and colleagues (Citation28) carried out a study in England linking a registry of patients hospitalized with an acute myocardial infarction to a primary care electronic health record used to identify patients with COPD and document use of beta-blockers after discharge from hospital. Beta-blocker use was associated with a halving of all-cause mortality (HR 0.50; 95% CI: 0.36–0.69). A proportion of this apparent benefit was due to considering subjects as exposed to beta-blockers or not, as of the first day of hospitalization when the drug could have started anytime during hospitalization, thus introducing immortal time. This is well illustrated by the Kaplan-Meir curves in of that paper (Citation28), where there is a major surge of deaths immediately after cohort entry in the nonusers of beta-blockers but the rate is gradual in the users (Citation28). In the sensitivity analysis based on follow-up was started from the day of discharge from hospital, thus excluding patients who died in hospital, the effect was attenuated (HR 0.64; 95% CI: 0.44–0.94). Since all patients included in this survivor cohort had to have beta-blocker use coded in the record, a less likely coding for those who died in hospital, this analysis is inherently subject to potential selection bias.
Figure 2. Forest plot of hazard and rate ratios of mortality associated with beta-blocker use in COPD from observational studies from and , and pooled estimates by a random effects approach, according to confounding and immortal time biases.
Figure 3. Forest plot of hazard and rate ratios of COPD exacerbation associated with beta-blocker use in COPD from observational studies from and , and pooled estimates by a random effects approach, according to confounding and immortal time biases.
lists the studies affected by immortal time bias. Predictably, the studies report significant reductions in mortality with beta-blocker use. The solution to these studies that introduced immortal time bias is to use a time-dependent definition of exposure that includes and properly classifies the exposure (Citation30).
Studies addressing biases
We found three studies that addressed these major biases using proper methods (Citation19, Citation32, Citation33). The first study, although classified as subject to confounding bias above, specifically recognized and avoided immortal time bias (Citation19). Ekström et al. used a cohort of patients starting long-term oxygen therapy for COPD in Sweden in 2005–2009 to assess the effect of beta-blocker use on mortality. It found no benefit of beta-blocker use on all-cause mortality (HR 1.19; 95% CI: 1.04–1.37). The approach employed a time-dependent definition for drug exposures, updating beta-blocker use over time, thus avoiding immortal time bias (Citation30). The time-dependent definition classified the patient as unexposed until their first beta-blocker prescription and exposure only started from that point onwards redefined every 90 days. This approach reduced the misclassification of exposure that leads to immortal time bias. On the other hand, some residual confounding could not be ruled out as the analysis was based on adjusting for baseline differences in age, sex, PaO2 air, PaCO2 air, BMI, WHO performance status and comorbidities (anemia, renal failure and cardiovascular diseases). There was no adjustment for prior exacerbations and the covariates were measured at baseline, not updated at the time of beta-blocker initiation.
A study which used a design that addresses confounding bias and avoids immortal time bias was initially conducted in one database and expanded to five databases from the USA, Italy and Taiwan (Citation32, Citation33). The initial study used the Taiwan claims database to identify a cohort of patients with a COPD diagnosis who initiated cardio-selective beta-blockers or calcium channel blockers (CCBs) during the study period 2007–2011 (Citation32). To control for confounding, 1:1 matching was used based on propensity scores that included, among many other risk factors, all respiratory medications and respiratory-related hospitalization, including for COPD exacerbation, bronchial asthma, pneumonia, influenza and acute bronchitis, during the baseline year prior to initiating the study drugs. The crude hazard ratio of death from any cause associated with beta-blocker use compared with CCBs was 0.54 (95% CI: 0.52–0.55) while, after matching on the propensity score, the hazard ratio increased to 0.85 (95% CI: 0.81–0.88). In a sensitivity analysis of the effect of potential residual confounding, the authors showed that if, even after matching, the proportion of severe COPD patients in the beta-blocker and CCB groups remains unbalanced at 5 and 10%, respectively, the hazard ratio would shift from 0.85 to 0.94 and would shift to 1.00 with larger differences.
In the extension of the study to five databases, the authors restricted the study population to patients with COPD hospitalized for acute coronary syndrome (ACS) and treated with beta-blockers or CCBs within 90 days of discharge (Citation33). After propensity score matching, the pooled hazard ratio of death from any cause associated with beta-blocker use compared with CCBs was 0.90 (95% CI: 0.78–1.02). For the outcome of COPD hospitalization, the pooled hazard ratio was 0.54 (95% CI: 0.47–0.61), which was unchanged when the analysis was restricted to the first 30 days of follow-up after treatment initiation (hazard ratio 0.55; 95% CI: 0.37–0.82). The fact that the hazard ratio of COPD hospitalization was the same within the first 30 days after treatment initiation, a biologically implausible effect in such a short exposure time, suggests that residual confounding remains important, as noted by the authors (Citation33).
Discussion
Observational studies are now used extensively to identify new indications for drugs prescribed in other indications, such as aspirin, an analgesic drug that was found to prevent myocardial infarction. In this paper, we evaluated methodological aspects of 18 observational studies of the effects of beta-blockers, drugs used for cardiovascular disease, on mortality and exacerbations in COPD. We found that most were subject to major bias from either confounding or immortal time, which tend to overestimate the benefit of a drug. We found two studies that addressed these biases in their design and data analysis, but concluded that residual confounding could not be ruled out. Thus the ongoing βLOCK-COPD trial of the beta-blocker metoprolol in patients with moderate to severe COPD will provide important evidence for this question (Citation10).
A recent re-analysis of the TORNADO trial of long-acting bronchodilators in COPD presents results that provide some reassurance as to the safety of beta-blockers in COPD (Citation6). Symptom and spirometric responses to therapy with long-acting beta-agonists and anticholinergics did not differ according to whether patients were receiving beta-blockers or not. A similar re-analysis of the SUMMIT trial also shows that the effects of long-acting beta-agonists in COPD on the rates of exacerbations, cardiovascular events and mortality are not affected by concomitant use of beta-blockers (Citation7).
There are several examples of medications used in other conditions that have been studied for the treatment of COPD. The case of inhaled corticosteroids, used mainly to treat asthma in the 1980s, was the object of several observational studies as a treatment for COPD, with results somewhat conflicting with those of the randomized trials (Citation34). Statins, used extensively to treat hypercholesterolemia, were identified as a potential drug for treating COPD on the basis of several observational studies that reported significant reductions in mortality and other major COPD adverse outcomes (Citation35,Citation36). However, the large STATCOPE randomized trial that evaluated this potential effectiveness of simvastatin in reducing the incidence of exacerbations in COPD was negative (Citation37).
The two sources of bias we identified in the 18 studies, from confounding and immortal time, are major. Confounding by contraindication is particularly important as beta-blockers have for a long time posed a dilemma for clinicians who treat COPD patients, a substantial number of whom have cardiovascular co-morbidity, because of the potential risk of bronchospasm and neutralization of the effectiveness of β2-agonists. Thus, beta-blocker prescription will tend to decrease with increasing COPD severity, which in turn is associated with higher mortality, as was observed in the Cooperative Cardiovascular Project re-analysis (Citation14). Indeed, the rate ratio of mortality adjusted for COPD severity was 0.89 (95% CI: 0.79–1.00), much different from the rate ratio of 0.60 (95% CI: 0.57–0.63) which was not adjusted for COPD prognostic factors and thus subject to bias from confounding. It is essential to account for such confounding by contraindication of beta-blocker use in COPD when estimating the effect of these drugs in the subgroup of patients with COPD. Yet, the two studies that included extensive data to control for COPD severity could not rule out residual confounding (Citation32, Citation33).
Immortal time bias, arising from the misclassification of exposure to beta-blockers and the exclusion of subjects, was also important in six studies. Unlike confounding bias that depends on accurate measures of COPD severity, immortal time bias is easily correctable as it does not require any additional data but rather a proper approach to data analysis that classifies exposure correctly over time. This bias has been identified in many observational studies that showed spectacular benefits with inhaled corticosteroids in COPD (Citation34). As well, several of the observational studies reporting significant protective effects of statins in COPD (Citation35, Citation36) were found to be affected by immortal time bias (Citation38). Indeed, the subsequent STATCOPE trial found no effect of simvastatin in reducing the incidence of exacerbations in COPD, confirming that the observational study results were due to bias (Citation37, Citation39).
In all, while observational studies are important to assess the effects of medications, their proper design and analysis are essential to avoid bias. Immortal time bias remains prevalent in the observational studies of several potential medications in COPD, including beta-blockers, but should be easily correctable. On the other hand, confounding by contraindication, which is rather unique and major in the case of beta-blockers and COPD, remains problematic and challenging. Thus, unlike the case of statins, the randomization in the ongoing βLOCK-COPD trial in COPD will be essential to eliminate these confounding concerns (Citation10).
Acknowledgments
Both authors contributed to the conception and writing of the manuscript.
Disclosure statement
S. Suissa has participated in advisory boards, as speaker, or received funding from AstraZeneca, Boehringer Ingelheim and Novartis. P. Ernst has no conflict to declare.
References
- Sin DD, Man S. A curious case of beta-blockers in chronic obstructive pulmonary disease. Arch Intern Med. 2010;170(10):849–850.
- Rabe KF, Hurst JR, Suissa S. Cardiovascular disease and COPD: dangerous liaisons? Eur Respir Rev. 2018;27(149).pii: 180057. doi:10.1183/16000617.0057-2018.
- DiNicolantonio JJ, Fares H, Niazi AK, Chatterjee S, D'Ascenzo F, Cerrato E, Biondi-Zoccai G, Lavie CJ, Bell DS, O'Keefe JH. β-Blockers in hypertension, diabetes, heart failure and acute myocardial infarction: a review of the literature. Open Heart. 2015;2(1):e000230. doi:10.1136/openhrt-2014-000230.
- Kotecha D, Manzano L, Krum H, Rosano G, Holmes J, Altman DG, Collins PD, Packer M, Wikstrand J, Coats AJ, et al. Effect of age and sex on efficacy and tolerability of β blockers in patients with heart failure with reduced ejection fraction: individual patient data meta-analysis. BMJ. 2016;353:i1855.
- Lipworth B, Wedzicha J, Devereux G, Vestbo J, Dransfield MT. Beta-blockers in COPD: time for reappraisal. Eur Respir J. 2016;48(3):880–888.
- Maltais F, Buhl R, Koch A, Amatto VC, Reid J, Grõnke L, Bothner U, Voß F, McGarvery L, Ferguson GT. β-Blockers in COPD: a cohort study from the TONADO research program. CHEST. 2018;153(6):1315–1325.
- Dransfield MT, McAllister DA, Anderson JA, Brook RD, Calverley PMA, Celli BR, Crim C, Gallot N, Martinez FJ, Scanlon PD, et al. β-blocker therapy and clinical outcomes in patients with moderate chronic obstructive pulmonary disease and heightened cardiovascular risk. An observational substudy of SUMMIT. Ann Am Thorac Soc. 2018;15(5):608–614.
- Etminan M, Jafari S, Carleton B, FitzGerald JM. Beta-blocker use and COPD mortality: a systematic review and meta-analysis. BMC Pulm Med. 2012;12(1):48.
- Du Q, Sun Y, Ding N, Lu L, Chen Y. Beta-blockers reduced the risk of mortality and exacerbation in patients with COPD: a meta-analysis of observational studies. PLOS One. 2014;9(11):e113048.
- Bhatt SP, Connett JE, Voelker H, Lindberg SM, Westfall E, Wells JM, Lazarus SC, Criner GJ, Dransfield MT. β-Blockers for the prevention of acute exacerbations of chronic obstructive pulmonary disease (βLOCK COPD): a randomised controlled study protocol. BMJ Open. 2016;6(6). e012292. doi:10.1136/bmjopen-2016-012292.
- Huang YL, Lai CC, Wang YH, Wang CY, Wang JY, Wang HC, Yu CJ, Chen L. Impact of selective and nonselective beta-blockers on the risk of severe exacerbations in patients with COPD. Int J Chron Obstruct Pulmon Dis. 2017;12:2987–2996.
- Liao K-M, Lin T-Y, Huang Y-B, Kuo C-C, Chen C-Y. The evaluation of β-adrenoceptor blocking agents in patients with COPD and congestive heart failure: a nationwide study. Int J Chron Obstruct Pulmon Dis. 2017;12:2573–2581. doi:10.2147/COPD.S141694.
- Gottlieb SS, McCarter RJ, Vogel RA. Effect of beta-blockade on mortality among high-risk and low-risk patients after myocardial infarction. N Engl J Med. 1998;339(8):489–497. doi:10.1056/NEJM199808203390801.
- Chen J, Radford MJ, Wang Y, Marciniak TA, Krumholz HM. Effectiveness of beta-blocker therapy after acute myocardial infarction in elderly patients with chronic obstructive pulmonary disease or asthma. J Am Coll Cardiol. 2001;37(7):1950–1956.
- Au DH, Bryson CL, Fan VS, Udris EM, Curtis JR, McDonell MB, Fihn SD. Beta-blockers as single-agent therapy for hypertension and the risk of mortality among patients with chronic obstructive pulmonary disease. Am J Med. 2004;117(12):925–931.
- van Gestel YR, Hoeks SE, Sin DD, Welten GM, Schouten O, Witteveen HJ, Simsek C, Stam H, Mertens FW, Bax JJ, et al. Impact of cardioselective beta-blockers on mortality in patients with chronic obstructive pulmonary disease and atherosclerosis. Am J Respir Crit Care Med. 2008;178(7):695–700.
- Hawkins NM, Huang Z, Pieper KS, Solomon SD, Kober L, Velazquez EJ, Swedberg K, Pfeffer MA, McMurray JJ, Maggioni AP, et al. Chronic obstructive pulmonary disease is an independent predictor of death but not atherosclerotic events in patients with myocardial infarction: analysis of the Valsartan in Acute Myocardial Infarction Trial (VALIANT). Eur J Heart Fail. 2009;11(3):292–298. doi:10.1093/eurjhf/hfp001.
- Mentz RJ, Wojdyla D, Fiuzat M, Chiswell K, Fonarow GC, O'Connor CM. Association of beta-blocker use and selectivity with outcomes in patients with heart failure and chronic obstructive pulmonary disease (from OPTIMIZE-HF). Am J Cardiol. 2013;111(4):582–587. doi:10.1016/j.amjcard.2012.10.041.
- Ekström MP, Hermansson AB, Ström KE. Effects of cardiovascular drugs on mortality in severe chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2013;187(7):715–720.
- Bhatt SP, Wells JM, Kinney GL, Washko GR Jr, Budoff M, Kim YI, Bailey WC, Nath H, Hokanson JE, Silverman EK, et al. β-Blockers are associated with a reduction in COPD exacerbations. Thorax. 2016;71(1):8–14.
- Duffy S, Marron R, Voelker H, Albert R, Connett J, Bailey W, Casaburi R, Cooper JA Jr, Curtis JL, Dransfield M, et al. Effect of beta-blockers on exacerbation rate and lung function in chronic obstructive pulmonary disease (COPD). Respir Res. 2017;18(1):124.
- Coiro S, Girerd N, Rossignol P, Ferreira JP, Maggioni A, Pitt B, Tritto I, Ambrosio G, Dickstein K, Zannad F. Association of beta-blocker treatment with mortality following myocardial infarction in patients with chronic obstructive pulmonary disease and heart failure or left ventricular dysfunction: a propensity matched-cohort analysis from the High-Risk Myocardial Infarction Database Initiative. Eur J Heart Fail. 2017;19(2):271–279.
- Sin DD, McAlister FA. The effects of beta-blockers on morbidity and mortality in a population-based cohort of 11,942 elderly patients with heart failure. Am J Med. 2002;113(8):650–656.
- Dransfield MT, Rowe SM, Johnson JE, Bailey WC, Gerald LB. Use of beta blockers and the risk of death in hospitalised patients with acute exacerbations of COPD. Thorax. 2008;63(4):301–305.
- Olenchock BA, Fonarow GG, Pan W, Hernandez A, Cannon CP. Current use of beta blockers in patients with reactive airway disease who are hospitalized with acute coronary syndromes. Am J Cardiol. 2009;103(3):295–300.
- Rutten FH, Zuithoff NP, Hak E, Grobbee DE, Hoes AW. Beta-blockers may reduce mortality and risk of exacerbations in patients with chronic obstructive pulmonary disease. Arch Intern Med. 2010;170(10):880–887.
- Short PM, Lipworth SI, Elder DH, Schembri S, Lipworth BJ. Effect of beta blockers in treatment of chronic obstructive pulmonary disease: a retrospective cohort study. BMJ. 2011;342:d2549.
- Quint JK, Herrett E, Bhaskaran K, Timmis A, Hemingway H, Wedzicha JA, Smeeth L. Effect of β blockers on mortality after myocardial infarction in adults with COPD: population based cohort study of UK electronic healthcare records. BMJ. 2013;347:f6650.
- Suissa S. Immortal time bias in observational studies of drug effects. Pharmacoepidemiol Drug Saf. 2007;16(3):241–249.
- Suissa S. Immortal time bias in pharmacoepidemiology. Am J Epidemiol. 2008;167(4):492–499.
- Suissa S, Ernst P. Biases in the observational study of beta blockers in COPD. Thorax. 2008;63(11):1026–1027.
- Dong Y-H, Chang C-H, Wu L-C, Lai M-S. Use of cardioselective β-blockers and overall death and cardiovascular outcomes in patients with COPD: a population-based cohort study. Eur J Clin Pharmacol. 2016;72(10):1265–1273. doi:10.1007/s00228-016-2097-y.
- Dong YH, Alcusky M, Maio V, Liu J, Liu M, Wu LC, Chang CH, Lai MS, Gagne JJ. Evidence of potential bias in a comparison of β blockers and calcium channel blockers in patients with chronic obstructive pulmonary disease and acute coronary syndrome: results of a multinational study. BMJ Open. 2017;7(3):e012997. doi:10.1136/bmjopen-2016-012997.
- Suissa S, Ernst P. Observational studies of inhaled corticosteroid effectiveness in COPD: lessons learned. CHEST. 2018;154(2):257–265.
- Dobler CC, Wong KK, Marks GB. Associations between statins and COPD: a systematic review. BMC Pulm Med. 2009;9(1):32.
- Horita N, Miyazawa N, Kojima R, Inoue M, Ishigatsubo Y, Ueda A, Kaneko T. Statins reduce all-cause mortality in chronic obstructive pulmonary disease: a systematic review and meta-analysis of observational studies. Respir Res. 2014;15(1):80.
- Criner GJ, Connett JE, Aaron SD, Albert RK, Bailey WC, Casaburi R, Cooper JA Jr, Curtis JL, Dransfield MT, Han MK, et al. Simvastatin for the prevention of exacerbations in moderate-to-severe COPD. N Engl J Med. 2014;370(23):2201–2210.
- Suissa S. Co-morbidity in COPD: the effects of cardiovascular drug therapies. Respiration. 2010;80(1):3–7.
- Suissa S. Simvastatin in moderate-to-severe COPD. N Engl J Med. 2014;371(10):969–970.