In the previous guidelines, chronic cough was considered merely as a symptom, or a consequence, of another disease (or causative) condition [678]. However, many disease conditions previously classified as “causes,” such as asthma, rhinitis, reflux or drugs, may not directly cause cough but act as “triggers” and sensitize the cough reflex [9]. It is also apparent that the majority of patients with the disease triad do not complain of troublesome coughs [9]. In addition, systematic reviews of randomized controlled trials (RCTs) have indicated that treatment responses to specific therapies, such as antiacid drugs or inhaled corticosteroids (ICS), are not solely therapeutic benefits but may include considerable placebo effects or self-remission [1011]. As the causal determination relies on the treatment response, it is possible that the proportion of cough previously explained by these diseases may have been overestimated. Thus, in the new guidelines, the conditions have been re-defined as “phenotypes” and translated into “treatable traits” of chronic cough [1] (Fig. 1).

Controlling treatable traits is highly important and should be attempted before considering antitussive drugs. A thorough clinical assessment, including a detailed history taking and physical examination, should help to identify which of the treatable traits underlies the patient's chronic cough (Fig. 2). However, the identification is frequently difficult, particularly in primary and secondary clinics, because there is still no test with good validity and practicability for guiding a trait-specific treatment. Induced sputum tests are considered the standard for predicting corticosteroid treatment responses or asthmatic cough [12], but they are not available in most clinics. The utility of methacholine challenge test is unclear in guiding the treatment strategy for cough patients without wheezing or breathlessness. Simple and convenient tests, such as fractional exhaled nitric oxide or blood eosinophil counts, have gained much attention for their predictability of the anti-inflammatory treatment response [13]; however, there is still no convincing evidence to advocate or refute their use in patients with chronic cough. Thus, sequential trials of each drug in turn are advocated in the new guidelines. However, antiacid drugs are not recommended unless patients have peptic symptoms or evidence of acid reflux, since nonacid reflux appears to be a major precipitant of the cough. Patient's clinical information and regional epidemiological data may help to determine the order in sequential drug trials. It is important to withdraw the trial if there is no treatment response or if the patient is revealed as not having a trait that responds to the treatment. The length of each drug trial depends on the pharmacology and clinical data reported in RCTs (for example, 1 month for ICS [11]). If the trial is successful, the treatment may be continued for several months to induce resolution of neuronal hypersensitivity.

Emphasis has been put on “cough control” interventions as well. Clinical recommendations have been drawn up for a few drugs, such as opiates, gabapentin or pregabalin; however, as their benefits have been empirically identified (they were not originally developed to treat cough; but they are mostly for pain), response rates are <50% and there are concerns about side effects. Thus, the antitussives should be carefully trialed only in patients with chronic refractory cough. Novel antitussives, such as P2X3 antagonists [14], are expected to be beneficial additions to our treatment strategy soon. Clinical recommendation has been also formulated for non-pharmacological cough control or speech language therapy, which is effective and safe when undertaken by experienced practitioners [15].

The guidelines also narratively address clinical questions that may not be easily addressed in RCTs or cannot be formulated as PICO (patient, intervention, comparison, outcomes) questions, including disease pathophysiology, cough assessment tools, diagnostic tests, or smoking cessation. Visual analogue scale (or simply asking “score your cough severity out of 10”) is recommended as the practical tool for cough assessment in daily clinic. Validated questionnaires, such as the Hull Airway Reflux Questionnaire [16] (please see http://issc.info for multiple lingual versions) or Reflux Symptom Index [17], are also recommended for the clinical diagnosis of nonacid airway reflux. Clinical pitfall regarding smoking cessation is introduced that the intervention may lead to transient increase in cough within the first month [1819]. Hopefully, the information will facilitate clinical decisions and patient consultations.

There are still many knowledge gaps in the guidelines [1]. First, most of clinical recommendations rely on only a few RCTs, thus warranting further clinical trials. Furthermore, the characteristics of cough have been poorly described in most of previous trials. High-quality RCTs are currently being conducted using validated outcomes, which we expect will help in estimating more precisely true treatment effects. Second, there is no practical and valid biomarker in use to guide a trait-specific therapy. Thus, we conclude that sequential therapeutic trials are still inevitable; further efforts should be made to develop and validate biomarkers to target specific traits in patients with chronic cough. Third, diagnostic and therapeutic pathways are not standardized but vary, even among cough specialists. The recently initiated ERS Clinical Research Collaboration, the NEUROCOUGH, will help to standardize patient care [20]. We hope that the new ERS cough guidelines will be a clinical framework to guide clinical thinking, develop further clinical trials and refine treatment strategies for patients with chronic cough in Europe and the Asia-Pacific region. As there will be local issues specific to each Asia-Pacific country [21], development of regional evidence and adaptation to local guidelines are encouraged.