LAUSR

We are grateful for the opportunity to respond to a letter to the editor regarding our manuscript “Concordant valid‐ ity of a digital peak cough flow meter to assess voluntary cough strength in individuals with ALS” [1, 2]. We thank Dr. Kulnik for his interest in our recent study that examined concordance of a handheld peak flow meter to assess peak expiratory flow during voluntary cough production com‐ pared to a laboratory pneumotachograph device in individu‐ als with ALS. Measurable expiratory parameters of voluntary cough airflow have been noted to distinguish aspiration status dur‐ ing swallowing in individuals with Parkinson’s disease [3], stroke [4], ALS [5] and head and neck cancer [6]. There‐ fore, measurement of peak expiratory flow during voluntary cough in populations at high risk for aspiration has been recommended to objectively index an individuals’ physi‐ ologic capacity to manage secretions and expel tracheal aspirate [7–9]. Given the inherent measurement capability, portability, low cost and means of obtaining an objective index of peak expiratory flow in real time; peak flow devices have been suggested and utilized in the clinical assessment of voluntary cough [8, 10–14]. Further, the recent Euro‐ pean Respiratory Society Statement recommended use of handheld peak flow devices to measure expiratory flow rates during voluntary cough [15]. We acknowledge and agree with Dr. Kulnik that the Microlife PF100 device examined in our study was designed to measure peak expiratory flow and forced expiratory volume in the first second of a forced exhalation. Our intent was not to suggest that the Microlife PF100 device was specifically designed for the assessment of peak expiratory flow during cough, but rather to deter‐ mine if this device could serve as a clinical surrogate for the impractical and expensive laboratory pneumotachograph method. Similar to previous authors and reports [6, 8, 9, 11] we referred to the digital device as a “peak cough flow meter” given it was measuring peak expiratory flow during a voluntary cough maneuver. However, Dr. Kulnik’s comment is well taken and a more precise terminology would have been to refer to the device as a “peak expiratory flow meter.” We conducted and reported results of a Pearson’s corre‐ lation coefficients and Lin’s concordance correlation coef‐ ficient (CCC), a statistical approach that has been previously utilized in similar studies [16–20]. In response to Dr. Kulnik we have additionally performed a Bland–Altman analysis here to estimate mean differences, bias and 95% limits of agreement for peak expiratory flow obtained during volun‐ tary cough on each device here. Figure 1 illustrates raw dif‐ ference in obtained peak flow rate between devices in every participant. Consistent with previous reports, individuals whose peak expiratory flow during cough was severely impaired and specifically less than 270 L/min demonstrated larger discrepancies between devices, with a trend for the digital device to overestimate peak expiratory flow at lower ranges compared to the laboratory pneumotachograph [21]. Given these results, the Bland–Altman analysis was con‐ ducted in both the full sample and for values obtained > 270 L/min using MedCalc Statistical Software (version 19.2.1; MedCalc Software Ltd, Ostend, Belgium). These results are * L. Tabor Gray Lauren.tabor@holy‐cross.com