Cough is among the most common complaints for which medical attention is sought1 yet, historically, despite its significance as a health issue worldwide, cough has been under-represented in terms of basic and clinical research efforts aimed at understanding its mechanisms and developing effective therapeutic agents. Thankfully, the past decade has witnessed an emergence of interest in cough within the scientific community. Highlighting the perceived importance of cough as a clinical problem, no less than five major pulmonary societies have published guidelines addressing the management of cough: the European Respiratory Society,2 the German Respiratory Society,3 the American College of Chest Physicians,4 the British Thoracic Society5 and the Japanese Respiratory Society.6 In addition, a task force of the European Respiratory Society published the first ever guidelines on the assessment of cough,7 aimed at promoting and improving the quality of cough-related research. Furthermore, the discovery of several pharmacological receptors relevant to the cough reflex—such as TRPV1,8 NOP19 and cannabinoid receptors10—has stimulated numerous investigative programmes within the pharmaceutical industry.

Research efforts in cough continue to be hindered by a dearth of clinically relevant, well validated measurement tools. Although cough reflex sensitivity can be accurately and reproducibly determined using inhalational (capsaicin or citric acid) cough challenge methodology, results generated in the laboratory may not correlate with the clinical state.11 For example, a potential antitussive agent under investigation may demonstrate the ability to inhibit experimentally-induced cough but fail to suppress pathological cough.11

Subjective cough-specific instruments such as the Leicester Cough Questionnaire (LCQ)12 and the Cough-specific Quality of Life Questionnaire (CQLQ)13 have been published, validated and have served as useful tools in clinical cough research. However, factors such as mood, level of vigilance, symptoms associated with cough (pain, vomiting, incontinence, syncope) and placebo effect14 may influence an individual’s perception of cough severity and frequency, as well as the effect of a pharmacological intervention on these parameters. Indeed, studies have demonstrated that subjective measures of cough are only moderately related to objectively measured cough.15 Thus, the optimal use of subjective symptom assessment tools would appear to be in conjunction with objective cough measurement. The ideal objective cough monitor would be a portable, ambulatory, digital-recording device able to store at least 24 h of information that could then be analysed using software capable of recognising a cough with adequate sensitivity and specificity. Although several objective cough monitoring systems are currently under development, only one is commercially available at present.16

In this issue of Thorax, Kelsall and colleagues17 employ their objective cough monitoring system (Vitalojak, Vitalograph, UK) to investigate whether, among patients with chronic cough presenting to a tertiary referral clinic, women have higher cough rates than men (see page 393). Furthermore, they examined whether other predictors of objective cough frequency could be discerned.

It is not surprising that women were shown to have significantly higher cough rates, albeit only during the night. Women are consistently over-represented in specialty referral centres. Previous studies have shown that healthy women have a more sensitive cough reflex than that of healthy men.18 19 The present study confirms the results of a previous trial of patients with pathological cough, in which women had heightened cough sensitivity compared with men.20 Notably, though, in this study sex and cough reflex sensitivity (to citric acid) independently predicted higher cough rates, implying that factors other than cough reflex sensitivity (and still to be identified) may be contributing to a greater amount of coughing in women. The inability of the LCQ to demonstrate a significantly poorer cough-related quality of life in women, despite a much higher cough rate than men, reminds us that a particular subjective instrument may omit information relevant to a particular study population. The ability of the CQLQ to discern sex-related differences21 22 probaby reflects its inclusion of items particularly distressful to women with chronic cough, such as urinary incontinence.

Remarkable in this study is the finding that older subjects had higher cough frequency than younger subjects. There appears to be a presumption in the literature that cough reflex sensitivity is diminished in elderly subjects because patients with a history of aspiration pneumonia, most of whom are elderly, have been shown to have a diminished response to citric acid.23 24 One study specifically evaluated cough reflex sensitivity to citric acid in a group of 110 subjects (60 men; age range 20–78 years) divided into groups from the third to eighth age decades. Cough reflex sensitivity did not differ significantly among the six age decades, suggesting that cough reflex sensitivity does not diminish with age.25 Regardless, the authors of the present study found that the relationship between cough frequency and age was independent of cough reflex sensitivity as well as duration of cough. Thus, an explanation for the increased cough frequency in older subjects in this study remains elusive (gastro-oesophageal reflux?), but provides fertile ground for further research.

Given the significance of cough as a clinical problem worldwide and the paucity of safe and effective antitussive drugs, high-quality investigation in the field of cough remains a priority. The optimal clinical study should include both subjective and objective cough-specific end points, measured in appropriately selected subject populations. The availability of practical, well-validated automated cough monitors will represent a major breakthrough in clinical cough research and is awaited with great anticipation.