Article Text

Download PDFPDF
S17 A relevant definition of cough bouts
  1. RJ Dockry1,
  2. KJ Holt1,
  3. JA Smith1,
  4. K McGuinness2
  1. 1University of Manchester, Manchester, UK
  2. 2Manchester NHS Foundation Trust, Manchester, UK


Introduction and Objectives Patients identify coughing ‘fits’ or ‘bouts’ as a key component of cough severity along with cough frequency, intensity and disruption of daily activities. Cough bouts have previously arbitrarily been defined as sequential coughs, each occurring within a 2-second window of the previous cough. With patterns of coughing potentially important in the interpretation of cough frequency data, a definition of bouts with clinical relevance is needed.

Methods A previously collected dataset of 24 hr recordings from refractory chronic cough patients with cough positions marked was re-analysed. Using a validated custom-written algorithm, cough bouts (≥2 coughs) were quantified using a range of increasing time intervals (0.5s, 1.0s, 1.5s, 2.0s, 2.5s and 3.0s) to define bouts within the individual datasets. The total number of bouts under each condition was then correlated with a cough severity visual analogue score (VAS), collected from the participant at the time of the recording, using a Spearman’s Rank Correlation. Single coughs were not included in the total number of bouts, but the influence of including these was assessed.

Results Cough data from 47 chronic cough patients was analysed. The correlation between the median number of individual cough sounds and VAS was r=0.31, p=0.04. The median number of bouts, and the correlation coefficient with the VAS are shown in figure 1. All correlations were significant (p<0.05), aside from the 1.5s window (p=0.06). The 3s window gave the best correlation coefficient (0.369, p=0.014). The inclusion of single coughs (added to the number of bouts) weakened the correlations for all cough bout definitions (all p>0.05).

Abstract S17 Figure 1

Correlation coefficient of total bouts and VAS, and median number of bouts calculated at each time window

Conclusions These data suggest that total cough bouts defined by longer time windows better reflect patient perceived cough severity. The commonly used 2s window may not be optimal if the goal is to reflect the impact of cough bouts on patients and further analysis of longer windows is required.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.