ABSTRACT
Background
The timing of LAMA administration, prescribed for morning dosing, has been questioned, particularly given the nocturnal symptoms many COPD patients experience. The LAMA BY NIGHT trial investigated the impact of LAMA dosing timing (morning vs. evening) on patient-related outcomes, and found no significant effect on risk of exacerbation, but a significant association with ICU admissions. This post hoc analysis of the LAMA BY NIGHT trial explored this association further according to adherence to administration time.
Objective
To determine whether the administration timing of LAMA affects the risk of ICU admission for COPD patients.
Methods
The LAMA BY NIGHT trial was a digital, multicenter, randomized, open-label study that enrolled 10,011 COPD patients. Adherence was measured at 6 and 12 months, with patients considered adherent if they used LAMA at least 6 days a week at the prescribed time. ICU admissions were tracked over 12-months, and relative risks (RR) were calculated to compare admission rates between morning and evening dosing groups. The analysis was stratified by adherence, to assess the sensitivity of the results.
Results
At 6 months, 6,747 patients (67%) completed follow-up questionnaires, of whom 80% were adherent. The risk of ICU admission was numerically lower in the evening dosing group among adherent patients (14/2,302 vs. 24/3,056, RR = 0.77, CI = (0.61–1.55), p = 0.55). At 12 months, ICU admissions were similar between groups (RR = 0.97, CI = (0.40–1.50), p = 1.00). Likewise, non-adherent patients showed no significant difference in ICU admissions between morning and evening dosing at 6 months or 12 months.
Conclusion
In this post hoc analysis of the LAMA BY NIGHT trial, evening administration of LAMA did not reduce the risk of ICU admissions among patients with COPD, and adherence to the assigned dosing schedule did not modify this association.
KEYWORDS: COPD, long-acting muscarinic antagonists (LAMA), randomized controlled trial, circadian variation, adherence, intensive care admission
Introduction
Treatment with long-acting muscarinic antagonists (LAMA) has become a mainstay in chronic obstructive pulmonary disease (COPD) management due to their ability to reduce exacerbation frequency and improve symptom control [1–3]. Although LAMAs are commonly prescribed to be taken once daily in the morning, the optimal timing of administration remains uncertain. Many COPD patients experience pronounced nocturnal symptoms, such as breathlessness and cough, that impair sleep and quality of life [4]. These night-time symptoms are predictive of higher mortality and more frequent exacerbations, highlighting the need for targeted therapeutic approaches that consider circadian variations [5]. Observational studies, including the STORICO study, have highlighted the circadian rhythm of COPD symptoms and suggest that treatment timing may need to be tailored to individual symptom pattern [6]. Furthermore, adherence to medication regimens is crucial for therapeutic efficacy. Previous clinical trials have demonstrated that non-adherence to therapy is associated with increased mortality and healthcare utilization across treatment arms, suggesting the presence of a potential ‘healthy adherer’ effect [7].
The LAMA BY NIGHT randomized controlled trial investigated whether the timing of LAMA administration (morning vs. evening) influenced clinical outcomes in patients with COPD, with primary endpoints including exacerbations and mortality [8]. The main trial found no significant differences in these primary outcomes between morning and evening dosing. However, a secondary observation suggested that evening LAMA administration might be associated with fewer ICU admissions, hinting at a potential benefit in preventing acute clinical deterioration in certain patients.
The present substudy specifically examines the role of adherence to LAMA timing in this context. While the main trial analyzed outcomes based solely on assigned dosing time, this analysis evaluates how patients’ actual adherence to the prescribed timing modifies the risk of ICU admission and other acute complications. This approach allows us to investigate whether the potential benefit of evening dosing is dependent on patients reliably taking LAMA at the intended time [9].
Therefore, this post hoc analysis of the LAMA BY NIGHT trial aimed to investigate whether evening administration of LAMA, compared with morning dosing, was associated with a reduced risk of ICU admission in patients with COPD, stratified by adherence to the assigned dosing regimen.
Methods
Study design and participants
The LAMA BY NIGHT trial [8] was a decentralized nationwide, randomized, open-label study designed to assess the effect of LAMA administration timing on clinical outcomes in patients with COPD. All Danish patients aged ≥30 years who had reimbursed a LAMA prescription by September 2022 were electronically invited to participate in the trial. National health registers and electronic questionnaires were used to collect follow-up data. Thus, no in person follow-up visits were required. The diagnosis of COPD and once daily use of LAMA were self-reported. The trial design is thoroughly described in the main study.
A total of 10,011 participants were randomly assigned to either morning or evening dosing. Adherence was assessed at 6 and 12 months using self-reported data, with adherence defined as using LAMA at least 6 days a week at the assigned time. Non-adherence was defined as taking LAMA at a different time than prescribed or using it less than 6 days a week. Participants who died during the study and were unable to complete the follow-up surveys were included in the adherent group to avoid introducing survival bias. However, participants who were alive but refrained from completing the follow-up surveys were excluded (Figure 1). Participants received automated electronic reminders monthly during the first 6 months and once at 9 months to reinforce dosing-time adherence and completion of follow-up questionnaires.
Figure 1.
Flowchart – inclusion and exclusion of participants in the sub study.
Main outcome
The primary outcome of interest was ICU admission, defined as at least one admission to ICU by any cause, during the follow-up period. ICU admissions were identified through linkage to national hospital registries using each participant’s unique identifier, ensuring complete and accurate capture of admission events. For the analysis, ICU admission rates were compared between participants randomized to morning versus evening LAMA dosing. These comparisons were further stratified by adherence status, as assessed through self-reported questionnaire data at 6 and 12 months, to evaluate whether adherence to the assigned dosing time modified the association with ICU admission.
Statistical analysis
Categorical variables were summarized using count and proportions, while mean and standard deviance (SD) were used for continuous covariates. The risk of ICU admission was compared between the morning and evening groups at both 6 and 12 months, stratified by adherence status. A p-value of <0.05 was considered statistically significant, with relative risks (RR) and confidence intervals (CI) used to quantify the effect of LAMA timing on ICU admissions. The difference was analyzed using chi-square tests with Yates continuity correction.
Statistical analyses were carried out using SAS software version 9.4 (SAS Institute, Cary, NC, USA) and R version 4.3.3 (R Foundation, Vienna, Austria). Relative risks were visualized using ggplot2 [10].
Results
Patient characteristics
At 6 months, the cohort consisted of 6,747 (67%) patients, thus excluding 3264 (33%) participants who were alive and not responding. In this cohort 5,396 (80%) were adherent and 1,351 (20%) were non-adherent. The baseline characteristics of patients in both the evening and morning LAMA groups were well balanced, with a mean age of 71 years and approximately equal proportions of males (53%). Across all groups, a slightly lower percentage of the patients were overweight and obese, and between 25% and 32% had a BMI greater than 30 kg/m2. GOLD classification revealed that most patients had moderate COPD (GOLD 2, 57.4%), followed by severe COPD (GOLD 3, 23.9%). Comorbidities were common, with a substantial number of patients presenting with a Charlson Comorbidity Index (CCI) score ≥5 (44.4%). The most common comorbidities were asthma (31%), diabetes (17%) and hypertension (54%) (Table 1).
Table 1.
Baseline characteristics.
| 6 months adherence |
6 months non-adherence |
|||
|---|---|---|---|---|
| Evening | Morning | Evening | Morning | |
| n | 2316 | 3080 | 976 | 375 |
| Sex = Male (%) | 1285 (55.5) | 1635 (53.1) | 508 (52.0) | 204 (54.4) |
| Age (mean ± SD) | 71.7 ± 8.2 | 71.8 ± 8.2 | 71.2 ± 8.3 | 70.9 ± 8.6 |
| BMI (%) | ||||
| <25 kg/m2 | 824 (35.6) | 1165 (37.8) | 363 (37.2) | 125 (33.3) |
| 25–30 kg/m2 | 844 (36.4) | 1139 (37.0) | 318 (32.6) | 132 (35.2) |
| >30 kg/m2 | 648 (28.0) | 776 (25.2) | 295 (30.2) | 118 (31.5) |
| GOLD-CLASS (%) | ||||
| 1 | 215 (14.2) | 257 (12.8) | 68 (11.1) | 38 (15.0) |
| 2 | 887 (58.8) | 1145 (57.1) | 336 (54.9) | 145 (57.3) |
| 3 | 314 (20.8) | 504 (25.1) | 173 (28.3) | 54 (21.3) |
| 4 | 93 (6.2) | 99 (4.9) | 35 (5.7) | 16 (6.3) |
| Asthma (%) | 745 (32.2) | 935 (30.4) | 280 (28.7) | 149 (39.7) |
| CCI | ||||
| 0–2 | 178 (7.7) | 225 (7.3) | 84 (8.6) | 34 (9.1) |
| 3–4 | 1114 (48.1) | 1494 (48.5) | 462 (47.3) | 163 (43.5) |
| ≥5 | 1024 (44.2) | 1361 (44.2) | 430 (44.1) | 178 (47.5) |
| Comorbidities | ||||
| Astma | 935 (30.4) | 745 (32.2) | 149 (39.7) | 280 (28.7) |
| Hypertension | 1643 (53.3) | 1288 (55.6) | 211 (56.3) | 494 (50.6) |
| Diabetes | 500 (16.2) | 411 (17.7) | 84 (22.4) | 156 (16.0) |
| ICS (%) | 1077 (46.5) | 1377 (44.7) | 430 (44.1) | 189 (50.4) |
| LABA (%) | 2158 (93.2) | 2890 (93.8) | 918 (94.1) | 355 (94.7) |
| MRC (%) | ||||
| 1 | 564 (24.4) | 674 (21.9) | 213 (21.8) | 74 (19.7) |
| 2 | 996 (43.0) | 1415 (45.9) | 412 (42.2) | 165 (44.0) |
| 3 | 492 (21.2) | 657 (21.3) | 232 (23.8) | 82 (21.9) |
| 4 | 220 (9.5) | 287 (9.3) | 102 (10.5) | 39 (10.4) |
| 5 | 44 (1.9) | 47 (1.5) | 17 (1.7) | 15 (4.0) |
Adherence assessment
At 6-months, 2,316 (43%) adherent patients were in the evening group and 3,080 (57%) were in the morning group. A majority of the non-adherent patients were in the evening group with 976 (72%) patients compared to 375 (28%) in the morning group. A substantial proportion of the study population did not respond to the questions regarding adherence, i.e. 3,264 (33%) at 6 months.
At 12 months, 5150 (52%) participants were excluded due to being alive and non-responding, leaving 4861 (48%) in the cohort. Adherence rates were similar, with 4204 (86%) adherent participants of which 1,747 (42%) were in the evening group and 2,457 (58%) in the morning group. The proportion of non-adherent patients decreased at 12 months with 657 (13%) being non-adherent. Hereof 516 (79%) were in the evening group and 141 (21%) in the morning group.
ICU admissions
Among adherent patients at 6 months, 14 of 2,302 (0.6%) participants experienced ICU admissions in the evening group compared with 24 of 3,056 (0.8%) participants in the morning group (RR = 0.77, CI = (0.40–1.50), p = 0.55). At 12 months, 29 of 1,718 (1.7%) ICU admissions occurred in the evening group compared with 42 of 2,415 (1.7%) in the morning group (RR = 0.97, CI = (0.61–1.55), p = 1.00) (Table 2, Figure 2).
Table 2.
ICU admissions according to adherence status.
| Risk of ICU admission | |||||
|---|---|---|---|---|---|
| No ICU admission | ICU admission | Relative Risk | Confidence interval | p-value | |
| 6 months adherence | |||||
| Morning | 3,056 | 24 (0.8%) | Ref. | Ref. | Ref. |
| Evening | 2,302 | 14 (0.6%) | 0.77 | (0.40–1.50) | 0.55 |
| 12 months adherence | |||||
| Morning | 2,415 | 42 (1.7%) | Ref. | Ref. | Ref. |
| Evening | 1,718 | 29 (1.7%) | 0.97 | (0.61–1.55) | 1 |
| 6 months non-adherence | |||||
| Morning | 372 | 3 (0.8%) | Ref. | Ref. | Ref. |
| Evening | 973 | 3 (0.3%) | 0.38 | (0.8–1.90) | 0.45 |
| 12 months non-adherence | |||||
| Morning | 515 | 1 (0.2%) | Ref. | Ref. | Ref. |
| Evening | 140 | 1 (0.7%) | 0.27 | (0.02–4.34) | 0.90 |
Figure 2.
Relative risk of ICU admission: Evening LAMA versus Morning LAMA.
Among non-adherent patients, ICU admissions within 6 months occurred in 3 out of 973 (0.3%) participants in the evening group compared to 3 of 372 (0.8%) in the morning group (RR = 0.38, CI = (0.08–1.90), p = 0.45). At 12 months, ICU admissions were reported in 1 of 140 (0.7%) participants in the evening group and 1 of 515 (0.2%) in the morning group (RR = 0.27, CI = (0.02–4.34), p = 0.90) (Table 2, Figure 2).
Discussion
In the present substudy, we found no statistically significant difference in the risk of ICU admission between evening and morning LAMA dosing, regardless of adherence status. Although point estimates consistently suggested a numerically lower risk of ICU admission in the evening group, particularly at 6 months, these differences were small and imprecise, and confidence intervals were wide. This pattern indicates that any potential effect of dosing time on the incidence of severe clinical deterioration is likely minimal. Our findings therefore complement the primary LAMA BY NIGHT results, which similarly showed no meaningful impact of dosing time on major clinical outcomes, including exacerbations and mortality. When placed in the context of existing literature, our findings align with randomized studies demonstrating no clinically relevant difference in lung function when long-acting bronchodilators are taken in the morning versus the evening [11]. While some studies of inhaled corticosteroids have reported modest improvements in trough FEV1 with evening dosing in asthma [12,13], these effects have been small, inconsistent, and do not appear to translate into differences in severe outcomes. Our results suggest that although circadian variations are a well-recognized feature of COPD symptoms and physiology, the timing of once-daily LAMA does not meaningfully influence the risk of ICU admission.
Adherence is an important factor when interpreting our results. Previous studies have demonstrated strong ‘healthy adherer’ effects in COPD, where adherent patients generally experience fewer exacerbations and hospitalizations [5,7]. In our analysis, however, adherence to dosing time did not modify the association between morning versus evening administration and ICU admission. This may indicate that timing adherence, unlike overall medication adherence, has limited impact on severe outcomes. Notably, adherence was lower in the evening dosing group, and a substantial proportion of patients did not complete follow-up questionnaires. Excluding these patients, who were alive but non-responding, may have introduced selection bias, although this approach was necessary within the constraints of the fully digital design.
The rarity of ICU admissions in this relatively moderate COPD population further limits the power to detect small differences between groups. Additionally, classification of deceased participants as adherent, implemented to minimize survival bias, may have increased the proportion of high-risk patients in the adherent category. Nevertheless, the consistency of our findings across adherence strata and timepoints strengthens the overall conclusion: timing of LAMA does not appear to influence the risk of ICU admission.
Taken together, our results suggest that dosing-time flexibility is safe, and that patients should prioritize taking their LAMA at a time that best fits their routine to maximize overall adherence, rather than adhering strictly to morning or evening dosing.
[13] In a study by Cushen et al. [5], the clinical impact of different adherence behaviors in severe COPD patients was explored, emphasizing the significant effect of adherence on exacerbation frequency and overall disease progression. This study found that patients who adhered to prescribed medication schedules, including the timing of their doses, had significantly fewer exacerbations and hospitalizations. In contrast, patients with poor adherence, particularly those with irregular dosing behaviors, had a higher frequency of acute exacerbations and a greater need for medical interventions, including ICU admissions.
A recent study by Matera et al. [14] explored the timing of long-acting bronchodilator administration and emphasized that nocturnal and early morning symptoms are common in COPD patients and have a significant impact on long-term outcomes. Once-daily LAMAs such as tiotropium and glycopyrronium offer sustained bronchodilation over 24 h, supporting flexible dosing schedules [15]. However, some studies have shown that evening administration can lead to more favorable trough FEV1 values the following morning, potentially offering better overnight symptom control and improved lung function at waking. This may be particularly relevant for patients experiencing early-morning breathlessness and nocturnal symptoms, aligning with the observed circadian variability in airway tone and airflow limitation in COPD [16,17].
Taken together, these studies suggest that both adherence and circadian symptom patterns can influence COPD outcomes. However, our results indicate that these mechanisms do not translate into differences in severe endpoints such as ICU admission. This discrepancy likely reflects that the clinical consequences of non-adherence identified by Cushen et al. relate primarily to overall medication use, not adherence to a specific dosing time. In our study, dosing-time adherence did not modify the association between timing and ICU admission, suggesting that the timing component itself carries limited physiological relevance for severe outcomes. Similarly, while Matera and others have shown that evening bronchodilator administration may improve short-term physiological markers such as trough FEV1 or overnight symptoms, these effects appear insufficient to influence rare, high-severity events like ICU admissions.
The main strength of this study lies in its randomized design and the innovative use of a digital recruitment strategy, which enabled the inclusion of over 10,000 patients with COPD – a scale unprecedented for this type of recruitment in COPD research. The digital setup facilitated complete follow-up on the primary outcome for a large subset of 6,747 patients, with no missing data, while minimizing participant burden. This approach allowed for efficient data collection without compromising quality, thereby providing robust evidence to address an important clinical question.
The lack of a statistically significant effect could be attributed to several factors. First, ICU admissions in COPD patients are relatively rare, particularly in those with moderate COPD, which limits the statistical power to detect meaningful differences. Given that the primary outcome was neutral, this highlights the challenge of identifying significant effects in a population with relatively low incidence of severe events.
Adherence to the prescribed timing was suboptimal in a substantial portion of patients, which may have diluted the potential benefit of evening dosing. Notably, non-adherence was more common in the evening group, with 976 patients reporting non-adherence compared to 375 in the morning group, potentially reflecting greater challenges in maintaining consistent evening routines. Number of patients alive and non-responding increased over time with 38% of the initial cohort not responding at 6 months follow-up survey and 52% 12-month follow-up survey. We suspect that patients who did not complete the follow-up surveys were not adherent to their trial allocation. While our adherence assessment was pragmatic and clearly defined, the exclusion of participants who were alive but did not complete follow-up surveys may have introduced selection bias. The excluded patients may have had a worsening of their COPD, inhibiting them from replying to the questionnaires. However, this approach was the most practical within the constraints of the trial’s fully digital design.
Interestingly, the point estimates indicated a numerically higher relative risk of ICU admission among adherent participants, suggesting a larger benefit form evening administration among non-adherent patients, although these findings were not statistically significant. This finding should be interpreted cautiously and likely reflects differences in underlying disease severity rather than a true adverse effect of adherence. Patients who are more symptomatic or have more severe COPD may be more motivated to follow their treatment schedule strictly, a form of confounding by indication, and thus carry a higher baseline risk of acute deterioration and ICU admission independent of adherence behavior. Furthermore, in the present analysis, deceased participants were classified as adherent to prevent survival bias, which may have unintentionally increased the proportion of high-risk patients in the adherent group. Finally, residual confounding related to unmeasured factors such as frailty, comorbidity burden, or health-seeking behavior could also contribute to this pattern. Collectively, these mechanisms may explain the observed numerical trend without implying a causal association between adherence and worse outcomes.
Furthermore, a considerable part of the participants refrained from completing the follow-up survey, which notably reduced the sample size. Future studies should increase incentives to complete follow-up surveys, to properly assess the risk of clinical outcomes across administration timing, as well as adherence [18]. Additionally, adherence was self-reported, which may introduce bias due to underreporting or recall bias. The study also predominantly included patients with moderate COPD which may limit generalizability to patients with more severe COPD who experience more frequent exacerbations and hospitalizations.
Conclusion
The timing of once-daily LAMA administration does not appear to influence the risk of ICU admission in patients with COPD, irrespective of adherence to the assigned dosing schedule. These findings suggest that treatment timing can be individualized according to patient preference, with emphasis on optimizing overall adherence rather than a specific dosing time.
Disclosure statement
N.D.J. reports personal fees from Sanofi. A.G.M reports personal fees from Sanofi and GSKJ.V. reports personal fees from GlaxoSmithKline, Chiesi Pharmaceuticals, Boehringer-Ingelheim, Novartis, and AstraZeneca.T.B.S. reports receiving grants from Sanofiduring the conduct of the study and grants from AstraZeneca, Bayer, Boston Scientific, GE Healthcare, GSK, Novartis, Novo Nordisk Pfizer, and Sanofi-Pasteur and personal fees from Amgen, AstraZeneca, Bayer, CSL Seqirus, GE Healthcare, GSK, IQVIA, Novartis, Parexel, and Sanofi outside the submitted work. No other disclo- sures were reported.
References
- [1].Wedzicha JA, Calverley PMA, Albert RK, et al. Prevention of COPD exacerbations: a European Respiratory Society/American Thoracic Society guideline. Eur Respir J. 2017;50(3):1602265. doi: 10.1183/13993003.02265-2016 [DOI] [PubMed] [Google Scholar]
- [2].Terry PD, Dhand R.. The 2023 GOLD Report: Updated Guidelines for Inhaled Pharmacological Therapy in Patients with Stable COPD. Pulm Ther. 2023;9(3):345–357. 10.1007/s41030-023-00233-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [3].Wedzicha JA, Miravitlles M, Hurst JR, et al. Management of COPD exacerbations: a European Respiratory Society/American Thoracic Society guideline. Eur Respir J. 2017;50(3):1602265. doi: 10.1183/13993003.02265-2016 [DOI] [PubMed] [Google Scholar]
- [4].Braghiroli A, Braido F, Piraino A, et al. Day and night control of COPD and role of pharmacotherapy: a review. Int J Chron Obstruct Pulmon Dis. 2020;15:1269–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Cushen B, Sulaiman I, Greene G, et al. The clinical impact of different adherence behaviors in patients with severe chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2018;197(12):1630–1633. doi: 10.1164/rccm.201712-2469LE [DOI] [PubMed] [Google Scholar]
- [6].Scichilone N, Antonelli Incalzi R, Blasi F, et al. Circadian rhythm of COPD symptoms in clinically based phenotypes. Results from the STORICO Italian observational study. BMC Pulm Med. 2019;19(1). doi: 10.1186/s12890-019-0935-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].Vestbo J, Anderson JA, Calverley PMA, et al. Adherence to inhaled therapy, mortality and hospital admission in COPD. Thorax. 2009;64(11):939–943. doi: 10.1136/thx.2009.113662 [DOI] [PubMed] [Google Scholar]
- [8].Sivapalan P, Rømer V, Wirenfeldt Klausen T, et al. Am/pm dosing of LAMA for COPD: a randomized controlled trial protocol using digital recruitment and registries. Front Med (Lausanne). 2024;11:11. doi: 10.3389/fmed.2024.1430169 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Haynes RB, Ackloo E, Sahota N, et al. Interventions for enhancing medication adherence. Cochrane database of systematic reviews. 2008. Apr 16. [DOI] [PubMed]
- [10].Wickham H. ggplot2. Cham: Springer International Publishing; 2016. [Google Scholar]
- [11].Magnussen H, Verkindre C, Jack D, et al. Indacaterol once-daily is equally effective dosed in the evening or morning in COPD. Respir Med. 2010;104(12):1869–1876. doi: 10.1016/j.rmed.2010.08.010 [DOI] [PubMed] [Google Scholar]
- [12].Postma DS, Sevette C, Martinat Y, et al. Treatment of asthma by the inhaled corticosteroid ciclesonide given either in the morning or evening. Eur Respir J. 2001;17(6):1083–1088. doi: 10.1183/09031936.01.00099701 [DOI] [PubMed] [Google Scholar]
- [13].Song JU, Park HK, Lee J. Impact of dosage timing of once-daily inhaled corticosteroids in asthma: a systematic review and meta-analysis. Ann Allergy Asthma Immunol. 2018;120(5):512–519. doi: 10.1016/j.anai.2017.12.021 [DOI] [PubMed] [Google Scholar]
- [14].Matera MG, Rinaldi B, Ambrosio C, et al. Is it preferable to administer a bronchodilator once- or twice-daily when treating COPD? Respir Med. 2023;219:107439. doi: 10.1016/j.rmed.2023.107439 [DOI] [PubMed] [Google Scholar]
- [15].Donohue JF, Fogarty C, Lötvall J, et al. Once-daily bronchodilators for chronic obstructive pulmonary disease: indacaterol versus tiotropium. Am J Respir Crit Care Med. 2010;182(2):155–162. doi: 10.1164/rccm.200910-1500OC [DOI] [PubMed] [Google Scholar]
- [16].Agusti A, Hedner J, Marin JM, et al. Night-time symptoms: a forgotten dimension of COPD. Eur Respir Rev. 2011;20(121):183–194. doi: 10.1183/09059180.00004311 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [17].Calverley PMA, Lee A, Towse L, et al. Effect of tiotropium bromide on circadian variation in airflow limitation in chronic obstructive pulmonary disease. Thorax. 2003;58(10):855–860. doi: 10.1136/thorax.58.10.855 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [18].Lassen MCH, Johansen ND, Vaduganathan M, et al. Electronically delivered nudges to increase influenza vaccination uptake in older adults with diabetes. JAMA Netw Open. 2023;6(12):e2347630. doi: 10.1001/jamanetworkopen.2023.47630 [DOI] [PMC free article] [PubMed] [Google Scholar]