The value of continuous cough monitoring: a narrative review

Introduction

Cough is one of the most common reasons patients seek medical care (1). In recent decades, there has been a growing appreciation of cough as both a symptom and as a disease itself, spurring a surge in cough-related research (2). Objective cough monitoring has become especially important in understanding respiratory diseases and in the development of antitussive therapies (3). In clinical trials for new cough treatments, measured objective cough frequency is often a primary endpoint (4-6). This heightened interest in cough has given rise to the concept of “coughomics”, which refers to the comprehensive, data-driven analysis of cough patterns in a manner analogous to genomics or proteomics (7). In this narrative review, we define continuous cough monitoring (CCM) as analytically validated acoustic detection and timestamping of cough events over extended periods of time (typically days to weeks and months) in real-world settings, as opposed to time-limited cough counting. Artificial intelligence (AI)-enabled CCM technologies are at the heart of coughomics, enabling the passive, unobtrusive, and privacy-preserving continuous collection of cough data as patients go about their daily lives. The thesis of this narrative review is that the ability of these technologies to automatically detect and count coughs with high accuracy, which transforms cough from a subjective symptom into a quantifiable digital biomarker, will improve our understanding of and management of cough (8-10). We present this article in accordance with the Narrative Review reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-876/rc).

Methods

In order to conduct a comprehensive narrative review to capture current knowledge on CCM, single-concept keyword searches were performed in databases including PubMed and Google Scholar (up to April 2025) using combinations of keywords such as “continuous cough monitoring”, “cough counts”, “cough frequency”, “longitudinal cough”, “wearable cough monitor”, and “cough monitoring clinical trial”. No MeSH terms or additional filters were applied in the search strategy. To be included, studies needed to involve objective cough measurement over prolonged periods (typically ≥24 hours of monitoring, often repeated or continuous over days/weeks) in human subjects, and to report how such data contributed to understanding or managing a health condition. Both device-based monitors (e.g., portable audio recorders, wearable sensors) and smartphone or ambient microphone approaches were included. We excluded studies that only assessed cough cross-sectionally (e.g., single-session cough sound analysis without longitudinal data) or relied purely on subjective cough reporting without any objective monitoring component. The search yielded a broad range of publications, which both reviewers screened for relevance to CCM, one unpublished data example (McGill University, 2024) is listed for cough monitoring insights in heart failure as these insights were presented at an online cough science event. In total, we incorporated findings from dozens of studies spanning chronic cough, respiratory infections, chronic lung diseases, and technological validation of cough monitoring tools (Table S1). This narrative synthesis aims to summarize the state-of-the-art in CCM and cough analytics, highlighting validated technologies and key clinical insights.

Cough quantification: from manual counts to AI algorithms

Historically, measuring cough frequency relied on labor-intensive methods. In the 1950s and 1960s, researchers made the first attempts to quantify coughs by using tape recorders triggered by sound to capture coughing episodes, which were later manually counted (11,12). These early systems were bulky and could only record intermittently, requiring patients to remain in a controlled environment. By the 1980s, the first rudimentary automated counters appeared, though they were limited by the technology of the time (13).

In the 2000s, more practical devices were developed. The Leicester Cough Monitor (LCM) and the VitaloJAK cough monitoring system were notable advances that allowed 24-hour ambulatory cough recording. These systems employed a wearable microphone to record audio, followed by offline analysis with either manual counting or semi-automated algorithms. Both LCM and VitaloJAK demonstrated good validity and became widely used in clinical research (14). However, these systems are limited in that they require patients to wear conspicuous equipment, record all ambient sounds that need to be processed by trained analysts, and are limited to 24-hour measurements.

Recent innovations have leveraged modern hardware and AI to enable CCM in real-world settings. High-performance microphones and mobile devices can now detect explosive cough-like sounds for months, and machine learning algorithms can automatically distinguish coughs from other sounds. This has led to a new generation of cough monitors that are fully automated, wearable or contactless, and capable of continuous 24/7 monitoring (15-18). Table 1 compares several modern cough monitoring platforms.

These systems represent the state-of-the-art in cough quantification technology, moving from the traditional 24-hour recordings to continuous, real-world monitoring. All have demonstrated an ability to detect cough events, though their performance and forms differ [wearable (15,16) vs. smartphone app (7,20) vs. contactless (17) vs. stethoscope-like chest patch (18,19)]. Notably, none currently has Food and Drug Administration (FDA) approval specifically for cough counting as a medical device (they are generally used under research or as audio recording devices) (21).

Critical to the correct use of cough monitors is a quantitative understanding of their accuracy in capturing cough as users go about their normal activities. This is generally assessed by comparing human annotations of cough in continuous recordings to the automated results from the monitor. These results can be compared on a cough-by-cough basis, yielding sensitivity and false positivity rates. However, since the output of interest to users is cough rates (expressed as coughs per hour) and not each cough, a preferred analysis is to compare the correlation of the hourly cough rates obtained by the two methods, for example, using Lin’s Concordance Correlation Coefficient (LCCC).

Another important consideration is the usability of the devices, measured either by qualitative surveys of users or more quantitatively in terms of the percentage of the study period in which the device is worn and counting coughs. Some of the devices are able to quantify the time during which the device is being worn by using the motion or photoplethysmography (PPG) sensors on the devices. Future studies are needed to compare adherence of various form factors such as watches versus pendants versus adhesives on the abdomen or chest, and how adherence is impacted by the user’s age and the duration of the study.

Use of CCM in specific diseases

CCM has begun to yield important clinical insights across a range of respiratory diseases. By capturing cough frequency objectively over days or weeks, CCM provides an unprecedented window into disease activity, treatment response, and patient well-being. Although a critical literature review is beyond the scope of this review, we provide a brief summary of some findings from key conditions.

Refractory chronic cough (RCC)

Cough lasting >8 weeks that is unexplained [unexplained chronic cough (UCC)] or unresponsive to therapy (RCC) is a condition with profound impact on quality of life (22). In a study by Chung et al., patients with problematic cough (but whose diagnosis was not ascertained) were monitored with a phone app for at least 30 days. There was significant inter- and intra-subject variability in cough. Of particular note, while some subjects had relatively consistent cough rates, the daily cough rates of about two-thirds of subjects varied substantially from day to day. Such variability means that in most individuals, a single day of observation often inaccurately reflects longer term trends in cough burden. For example, in subjects with high variability any given day had only a ~30% chance of reflecting their longer-term average (7). It remains unclear how this insight will be incorporated by the US FDA when they interpret clinical trials that only have 24-hour observations.

VitaloJAK 24-hour monitoring has been a primary endpoint employed in the development of multiple P2X3 receptor antagonist anti-tussive therapies. For example, gefapixant was evaluated in two Phase 3 trials (COUGH-1/2) (6), and eliapixant was tested in the PAGANINI trial (23). Gefapixant was not approved by the US FDA, and the eliapixant program was discontinued. More recently, an exploratory trial of a novel inhaled therapy—an alkaline hypertonic divalent salt aerosol (designed to hydrate the airway surface)—in RCC utilized continuous AI-based cough monitoring to measure efficacy (24). In that study, 3 weeks of continuous monitoring were divided into baseline, placebo, and treatment periods. The results, published in 2024, was the first example of early generation AI-powered cough counters being used to demonstrate drug efficacy in a randomized trial (24).

The ability to precisely quantify cough over >24 hours is also aiding the development of new classes of antitussives. Companies such as Nocion Therapeutics and Genentech have initiated trials of novel agents (e.g., a selective sodium channel blocker by Nocion [currently in a phase 2b trial], and a TRPA1 antagonist by Genentech [which is now discontinued]) for chronic cough, in which continuous cough counts are being used to assess efficacy (25). These studies reflect a broader trend: objective cough monitoring is becoming an important tool to augment subjective assessments and objectively evaluate treatment benefit in chronic cough. Continuous monitoring not only captures whether a therapy reduces cough rates, but also its impact on cough patterns such as time-of-day effects or progressive improvements over weeks that inform dosing and mechanism. Of particular interest is the impact of therapy on periods of intense coughing (termed bouts, epochs or spells) that patients find particularly problematic.

Chronic obstructive pulmonary disease (COPD)

Cough is a cardinal symptom of COPD and is associated with mucus hypersecretion and disease progression (26,27). CCM is providing new insight into the natural history of cough in COPD and its potential as an early indicator of exacerbations. Crooks et al. (2021) objectively measured cough counts in COPD patients during the convalescent period after an exacerbation and documented a clear reduction in cough frequency as the exacerbation resolved (28). Building on such findings, researchers have explored whether increases in cough frequency could serve as an early warning for impending exacerbations. Remarkably, cough-based alerts predicted 45% of COPD exacerbations, a mean of 3.4 days before onset, with a low false-alarm rate (about 1 false alert per 100 days) (28). Integrating cough count trends into a multi-parameter alert algorithm could improve early detection of COPD deteriorations.

Beyond exacerbation prediction, continuous cough data highlight the overall burden of disease in COPD. Patients with chronic COPD cough frequently, especially those with chronic bronchitis phenotypes. In stable COPD, cough frequency has been linked with worse lung function and mortality risk (29). Continuous monitors could quantify daily cough as an objective outcome in interventional studies—for example, to test if a therapy (such as a mucolytic or anti-inflammatory) reduces cough frequency in COPD patients over time. This objective measure might complement patient-reported outcomes.

Bronchiectasis

Persistent, productive cough due to airway damage and mucus accumulation is a protean manifestation of bronchiectasis. Cough frequency is an important indicator of disease activity in bronchiectasis and response to airway clearance therapies. CCM has recently been applied to quantify the effectiveness of airway clearance techniques like oscillatory positive expiratory pressure devices and high-frequency chest wall oscillation. In a pilot study at National Jewish Health, researchers used CCM to track cough frequency in bronchiectasis patients during their daily airway clearance regimen (which included devices such as the Aerobika^® oscillatory PEP, high-frequency chest oscillation vest, and the Volara^® lung expansion system). The initial findings showed that implementing a standardized airway clearance protocol led to a rapid reduction in coughing frequency in these patients, indicating successful mucus mobilization (30). Interestingly, the data suggested that different techniques had different impacts on cough - these granular insights were only possible through continuous monitoring during and after therapy sessions.

Congestive heart failure (CHF) and gastroesophageal reflux disease (GERD)

Chronic cough is not only linked to primary respiratory diseases but also to conditions like CHF and GERD (31). Patients with decompensated heart failure may develop a wet cough due to pulmonary edema, and that cough often improves with diuresis. There are anecdotal reports from continuous monitoring that illustrate this trajectory. For example, clinicians at McGill University observed that a heart failure patient’s cough counts tracked closely with their NT-proBNP levels: as aggressive diuretic therapy relieved fluid overload, daily cough counts fell in parallel, mirroring the improvement in cardiac function (unpublished data from McGill University, 2024). This suggests that CCM might serve as a noninvasive indicator of fluid status—a sudden uptick in cough frequency in a heart failure patient at home might signal pulmonary edema and prompt early intervention, even before overt symptoms like orthopnea worsen. Similarly, in GERD-related cough (a common phenotype of chronic cough), continuous monitoring can objectify the pattern (for instance, predominantly post-prandial cough bursts, or nocturnal cough when lying flat). Early case studies have noted that patients with acid reflux who initiate proton pump inhibitor therapy show a decline in cough frequency over weeks, and interestingly, the timing of cough episodes may shift (less coughing after meals once acid is suppressed). While these observations are so far anecdotal, they highlight potential novel uses of CCM: in CHF, to titrate therapy and monitor for relapse; in GERD, to evaluate adequacy of reflux control. At minimum, these “cough profiles” expand our understanding of how systemic conditions manifest in respiratory reflexes.

Tuberculosis (TB) and other infectious diseases

Cough is a hallmark of pulmonary TB and has major public health and epidemiological importance since it is often the trigger for a clinical evaluation and facilitates disease transmission. CCM has broken new ground in TB care by providing a quantitative measure of cough that can be used for research, diagnosis and treatment monitoring. A multi-country study by Huddart et al. (2023) demonstrated the feasibility of smartphone-based cough monitoring in resource-limited settings for TB patients (8). Participants with presumptive TB wore a smartphone running a cough-tracker app continuously for 14 days, and their cough patterns were analyzed. The study found that patients with microbiologically confirmed TB tended to start with very high cough rates that then decreased rapidly after initiation of proper anti-TB therapy, whereas those with non-TB diagnoses (pneumonia, asthma, etc.) presented with lower cough rates that did not decrease with anti-tuberculous therapy (8). Continuous monitoring might therefore be used to identify patients who are misdiagnosed or not responding to therapy and could benefit from regimen changes.

Beyond TB, other diseases with major global health impact cause cough and are being studied under the lens of coughomics. In malaria, fever is the protean manifestation but preliminary observations in malaria-endemic regions have noted that kids with malaria can present with significant cough, which when evaluated objectively can be as common in children with malaria as those with pneumonia (32).

While much work remains to be done, CCM can be thought of as a form of “acoustic epidemiology” (33), where the epidemiologic patterns of cough (frequency, diurnal variation, patterns of bouts, cough-free time) across different diseases may be informative and clinically useful.

Table 2 provides a summary of key findings from CCM studies across these respiratory and related conditions.

Coughomics: insights from CCM

Perhaps one of the most exciting advantages of CCM is the ability to study cough over prolonged periods of time in the general population as patients go about their daily routines. Passive, long-term monitoring is revealing new insights about what is “normal” for cough and how it varies between and within individuals - insights that have been termed by Chung et al as “coughomics” (34).

How much do people cough?

Traditional teaching is that healthy people do cough occasionally, but not nearly as much as those with respiratory disease. This has been confirmed by objective data. Studies using 24-hour recordings in healthy volunteers found that many healthy individuals cough only a handful of times per day. In one study, healthy subjects coughed a geometric mean of 4.6 times over 24 hours, with a wide range and variability from 0 to 136 coughs overall (35). By contrast, patients with chronic cough can cough hundreds of times per day (one early study of chronic coughers reported a median of ~270 coughs/24 h) (36).

Individual cough patterns

One of the striking findings from real-world data is the high intra-individual variability in cough frequency. Even in the absence of interventions, cough counts can fluctuate from day to day. As discussed above, the 30-day persistent cough study showed that some patients had relatively stable cough day-to-day, but most varied significantly (7). This variability raises the question: What is the optimal monitoring duration to capture a representative cough profile for an individual? Early analyses indicate that one day of monitoring is often not enough (7,37). For many patients, averaging multiple days provides a more reliable measure of their cough burden. One statistical simulation provides an approach to answering this question and suggests that, in general, 7 days is ideal for chronic cough patients to establish baseline and has the added advantage of taking into consideration a week cycle of activities which may impact coughing (38).

Subjective vs. objective cough assessment

Real-world monitoring has demonstrated the often poor correlation between patients’ perception of their coughing and the actual cough counts. In chronic cough studies, subjective ratings (such as visual analog scales or cough severity diaries) show only modest correlation with objective cough frequency (39). Lee et al. (2023) found that across a group of chronic cough patients, the correlation between 24-hour cough count and a daily symptom score was weak (Spearman’s rho ranging from 0.2 to 0.3). Some patients who reported severe cough had only moderate objective counts, and vice versa. Moreover, when looking at individual patients over time, the relationship varied widely—for some, their day-to-day cough count tracked closely with their self-rated severity, but in others the subjective and objective measures seemed unrelated. This mismatch may be due to multiple factors: individual differences in cough perception or tolerance, the influence of cough intensity (a few very violent coughs might feel worse than many mild throat clears), or simply recall bias. It underscores why objective monitoring can complement patient-reported outcomes. Both are important: frequency relates to physiological burden and can be counted, while severity encompasses the patient’s suffering and might depend on context (for example, a single cough during a quiet meeting might be rated as more bothersome than ten coughs while home alone). Real-world research is ongoing to better understand how patients perceive cough in daily life and how that maps to the data from their devices.

Average cough bout length and patterns

Analyzing continuous data also allows us to study periods of intense coughing or cough bouts. Clinically, patients often describe “coughing fits” rather than isolated coughs. Smith and colleagues have explored definitions of a cough bout using real-world 24-hour recordings (40). By trying different definitions (e.g., using varying intervals in seconds by which coughs were separated as different definitions), they found that defining a bout as any sequence of coughs spaced no more than ~3 seconds apart is reasonable and that single, isolated coughs should be considered separately. Using this definition, most bouts in chronic cough patients consist of a few coughs in succession (median bout length might be 2–3 coughs). Interestingly, the total number of cough bouts per day and the average bout length may correlate better with patient-rated severity than the sheer number of individual coughs. In practice, continuous monitoring data can be post-processed to calculate metrics like cough frequency (coughs/hour), cough bout frequency (bouts/hour), average bout duration, and even time spent coughing per hour. Among these, time spent coughing has been shown to correlate closely with the number of individual coughs and thus may be a particularly important metric (40).

Chronobiology of cough

Real-world cough monitoring is providing insights into when people cough. There is a clear circadian pattern in most people, though the ratio of day and bedtime coughing varies within and between people. Despite this, it is possible that nocturnal monitoring may be informative, as was shown in a subset of COPD and asthma patients, as reports that frequent overnight coughing is correlated with poor disease control (10). Additionally, seasonal and environmental influences could be observed. This points to a future where environmental health monitoring might integrate community cough data as an indicator of air quality effects. On an individual level, patients can review their own cough trends: some asthma patients using monitors have identified triggers such as cold weather days or high pollen days causing upticks in their cough, enabling them to adjust preventive measures (e.g., use inhalers pre-emptively on days they expect higher cough).

In summary, CCM in the real world has moved cough assessment from the qualitative to the quantitative realm. It establishes objective baselines for individuals, captures variability and patterns that short clinic visits cannot, and helps align treatment with actual patient needs. It also empowers patients with data that enables a more informed and empowered conversation with health care providers. These developments will only grow as smartphone and wearable-based apps make coughomics accessible to broad populations.

Future directions

The advent of CCM and “coughomics” is poised to transform respiratory research and clinical care in several ways. As technology and adoption advance, we anticipate:

Technical advances

Technology for CCM is rapidly being developed and validated. Clinical studies to compare and contrast the advantages and limitations of various devices, such as watches, pendants, and adhesives, in terms of accuracy and acceptability are needed. Progress is being made in understanding when the device is being worn and analysis of this data from clinical trials is providing insights into the factors that impact adherence with monitoring. Clarifying the ability of modern CCM technologies to distinguish between the user’s and others’ coughs, and the implications of this capability for device accuracy, will further assist researchers in choosing between different CCM technologies.

Cohort-specific studies and new disease applications

Thus far, much of the work has focused on chronic cough and COPD, but there is great interest in applying cough monitoring to other respiratory conditions. In particular, idiopathic pulmonary fibrosis (IPF) is a condition where cough is problematic and likely to be an informative biomarker and needs to be more rigorously studied. Up to 80% of IPF patients have a chronic dry cough that significantly affects the patient’s quality of life. Furthermore, cough frequency in IPF has been associated with disease progression and even mortality—patients who cough more have worse outcomes on average (41). In addition, cough is a key symptom of uncontrolled asthma, and objective monitoring might help detect early loss of control or assess responses to asthma treatments.

Integration into telehealth and digital therapeutics (DTx)

Continuous cough monitors are well-suited for integration into the growing field of digital telehealth. We foresee cough counting being incorporated into remote patient monitoring programs for conditions like COPD, asthma, heart failure, oncological therapy and post-surgery recovery. For instance, a connected cough sensor could alert a telehealth nurse that a COPD patient’s cough frequency has doubled from baseline, prompting a check-in call or a telemedicine visit to adjust therapy before a full-blown exacerbation occurs. Wearables like smartwatches and earbuds are being developed with microphones that could passively monitor cough in the background (42), turning consumer devices into health monitors. The transition this enables to proactive care could reduce hospitalizations. This also aligns with trends towards increased patient control of care as seen by some chronic cough patients using monitoring to conduct “n-of-1” trials, for example testing if a certain lozenge or breathing exercise reduces their cough count. As these systems become more user-friendly and validated, we predict they will become part of routine care for chronic respiratory illnesses, similar to how continuous glucose monitors are becoming the standard in diabetes care.

Work is ongoing to transform CCM from a research tool to an actual digital therapy. DTx has been developed for many indications such as insomnia, diabetes and depression. Because chronic cough is often a hypersensitivity condition that is amenable to behavioral suppression therapy, it is likely that integrating behavioral instructions and cough monitoring in digital interventions, delivered via applications, could reduce cough (43-45). CCM is a critical feature in such a DTx as it enables patients to view their progress over time (likely improving adherence), and underpins a “feedback loop” where treatment can be tailored in real time according to a patient’s cough patterns. However, further research is needed to define the impact of real time feedback with acceptable rates of false positive and negative coughs on the therapeutic value of such DTx. Recently, Hyfe and Kyorin Pharmaceutical Co., Ltd. announced a partnership to combine a digital curriculum and monitoring as a DTx for refractory and UCC (46). Such a DTx approach, for instance, via a smartphone app, could be used as a standalone therapy or as prescription drug use related software [PDURS (47)] to augment the impact of an antitussive medication.

Public health surveillance and epidemiology

On a population level, CCM could become a novel tool for public health surveillance of respiratory infections. Real-time aggregation of cough data from many individuals can potentially detect outbreaks or track disease prevalence. Research in acoustic epidemiology has suggested that monitoring trends in cough frequency (and even cough sound characteristics) in a community might allow early detection of infectious outbreaks (48). Another interesting avenue is monitoring environmental and occupational cough. By equipping workers in certain industries (miners, farmers, 911 first responders, etc.) with cough monitors, agencies could quantitatively assess exposure effects and intervene earlier for lung health. Over years, these data might reveal, for instance, that firefighters have subtle increases in night-time cough after heavy smoke exposure, correlating with developing bronchitis—information that could inform health guidance and policies.

Refinement of cough biomarkers and omics integration

As coughomics matures, it will likely integrate with other “omics” and biomarkers to provide a comprehensive picture of disease. For instance, cough frequency might be combined with sound analysis features (frequency spectrum of cough sound, cough intensity) to create a multidimensional cough profile. Early work has shown that features of cough sound (e.g., audio frequency patterns) might distinguish different etiologies. In the future, a continuous monitor might not only count coughs but also classify them (this cough sounds like a dry cough vs. a wet cough). This moves toward an era of diagnostic cough AI, where cough patterns serve as a non-invasive diagnostic aid. By combining cough data with genomic, proteomic, or microbiome data (for example, correlating cough trends with airway inflammatory biomarkers or sputum microbiology), researchers can gain deeper insights into disease mechanisms. Coughomics could also help endotype chronic cough patients: some might have a “neurogenic” profile (high cough frequency, normal lung function, high cough reflex sensitivity) while others have an “inflammatory” profile (moderate cough frequency but lots of sputum and airway inflammation). This could guide whether neuromodulators or anti-inflammatories are the better treatment approach.

Regulatory approval and clinical adoption

For cough monitoring to truly become part of standard care and drug development, formal regulatory pathways need to be cleared. As of April 2025, no cough monitoring system has FDA clearance specifically for cough counting (21). However, efforts are underway: manufacturers are pursuing FDA’s de novo device approval route for their cough-counting algorithms. Key to regulatory approval will be demonstrating clinical validity (that the cough count correlates with human annotation) and safety (ensuring data security and that the device doesn’t fail in a way that harms patients). The field is likely following in the footsteps of continuous glucose monitors or digital ECG patches, which gained acceptance after robust trials. Future asthma or COPD guidelines might include cough count as a criterion for control or as a trigger for stepping up therapy (“if cough frequency doubles from baseline, consider therapy adjustment”). Another aspect is standardization: international societies may put forth recommendations on how to perform cough monitoring (minimum duration, handling of data, definition of a cough event, etc.) so that data are comparable across studies. The establishment of reference ranges (normal cough counts by age, gender, etc.) and minimally important differences (how much reduction in cough count is clinically noticeable) are areas of active research (49).

In summary, the future of coughomics is bright. We anticipate an era where every cough tells a story—whether it’s a warning of an asthma flare, a metric of a drug’s efficacy, or a data point in a public health map. As technology continues to improve the barrier to adoption will lower. Cough, once relegated to the background of clinical observation, is now taking center stage as a rich source of data—and CCM is the tool that makes it possible.

Conclusions

CCM is reshaping our understanding of cough in both health and disease. By leveraging AI and ubiquitous sensing devices, we can now objectively quantify cough frequency and patterns over long periods, turning cough into a precise clinical metric. This narrative review highlights how CCM—the essence of “coughomics”—has advanced cough science and patient care. Modern cough monitors have demonstrated high accuracy in identifying coughs, enabling their use as digital endpoints in clinical trials and daily management. Across chronic respiratory diseases like RCC, COPD, and bronchiectasis, continuous monitoring has uncovered significant variability and provided novel insights: we can measure the true impact of a chronic cough, detect exacerbations early, and assess, in real time, whether a treatment is reducing a patient’s cough burden.

Importantly, CCM is not just a research tool; it is steadily moving into practical applications. Patients are beginning to use cough apps to track their condition, and physician scientists are testing how cough data can inform care decisions. With further validation and eventual regulatory approval, we are likely to see cough monitors integrated into routine practice, similar to how ambulatory heart monitors or glucose sensors are used today. The ability to monitor cough continuously fills a longstanding gap in respiratory medicine, where we have long relied on intermittent lung function tests or qualitative history. Now, cough—one of the body’s most vital protective mechanisms—can be monitored as closely as heart rate or blood pressure.

In drug development, CCM is already accelerating progress by providing objective, sensitive endpoints for new therapies. For patients, this means new treatments can be assessed more rigorously and brought to market with clear evidence of benefit. In clinical care, patients with chronic cough can have their condition quantified, helping to validate their experience and tailor therapies (for example, seeing improvement in numbers can encourage adherence and persistence with a treatment plan).

In conclusion, CCM is transforming cough from a subjective symptom into an objective vital sign. It enriches clinical insight, guides therapeutic interventions, and empowers both patients and providers with actionable data. As technology, research, and clinical practice continue to converge in the realm of cough monitoring, we can expect improved outcomes for patients suffering from cough and a deeper understanding of the role cough plays in human disease. Cough, often overlooked, is finally getting the scientific attention it deserves—and continuous monitoring is the key that has unlocked this new era of cough evaluation.

Acknowledgments

The authors used OpenAI’s ChatGPT o3 model with its DeepResearch capabilities on April 24th, 2025 to build on top of the provided draft of original manuscript (including sections as outlined in this narrative review) with a prompt: “You’re writing a paper on ‘The Value of Continuous Cough Monitoring’ and as part of this paper, doing a literature review on https://pubmed.ncbi.nlm.nih.gov/ and https://scholar.google.com/ related to keywords such as ‘cough monitoring’, ‘continuous cough monitoring’, ‘cough counts’, ‘cough frequency’, ‘longitudinal cough monitoring’ (and suggest more, if you see fit). Time range: no restriction. Focused on insights from continuous cough data on any condition. Prioritize insights from continuous cough data. Make it academic and professional, be specific and diligent. Use this draft manuscript as the guiding document” [attaching original draft manuscript outlining key bullet points and focus areas]. Authors are fully responsible for the content of this manuscript, as it was further manually reviewed and edited, and every reference double-checked and incorporated.

Footnote

Provenance and Peer Review: This article was commissioned by the Guest Editors (Kian Fan Chung and Woo-Jung Song) for the series “The Thirteenth London International Cough Symposium” published in Journal of Thoracic Disease. The article has undergone external peer review.

Reporting Checklist: The authors have completed the Narrative Review reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-876/rc

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-876/prf

Funding: None.

Conflicts of Interest: Both authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-876/coif). The series “The Thirteenth London International Cough Symposium” was commissioned by the editorial office without any funding or sponsorship. M.G. and P.M.S. are employees and stock holders of Hyfe Inc. The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Finley CR, Chan DS, Garrison S, et al. What are the most common conditions in primary care? Systematic review. Can Fam Physician 2018;64:832-40.
2. Morice A, Dicpinigaitis P, McGarvey L, et al. Chronic cough: new insights and future prospects. Eur Respir Rev 2021;30:210127. [Crossref] [PubMed]
3. Turner RD, Birring SS. Measuring cough: what really matters? J Thorac Dis 2023;15:2288-99. [Crossref] [PubMed]
4. Smith JA, Morice AH, Birring SS, et al. Camlipixant in Refractory Chronic Cough: A Phase 2a, Randomized Controlled Trial (RELIEF). Am J Respir Crit Care Med 2025;211:1072-5. [Crossref] [PubMed]
5. Maher TM, Avram C, Bortey E, et al. Nalbuphine Tablets for Cough in Patients with Idiopathic Pulmonary Fibrosis. NEJM Evid 2023;2:EVIDoa2300083.
6. McGarvey LP, Birring SS, Morice AH, et al. Efficacy and safety of gefapixant, a P2X(3) receptor antagonist, in refractory chronic cough and unexplained chronic cough (COUGH-1 and COUGH-2): results from two double-blind, randomised, parallel-group, placebo-controlled, phase 3 trials. Lancet 2022;399:909-23. [Crossref] [PubMed]
7. Chung KF, Chaccour C, Jover L, et al. Longitudinal Cough Frequency Monitoring in Persistent Coughers: Daily Variability and Predictability. Lung 2024;202:561-8. [Crossref] [PubMed]
8. Huddart S, Asege L, Jaganath D, et al. Continuous cough monitoring: a novel digital biomarker for TB diagnosis and treatment response monitoring. Int J Tuberc Lung Dis 2023;27:221-2. [Crossref] [PubMed]
9. Altshuler E, Tannir B, Jolicoeur G, et al. Digital cough monitoring - A potential predictive acoustic biomarker of clinical outcomes in hospitalized COVID-19 patients. J Biomed Inform 2023;138:104283. [Crossref] [PubMed]
10. Rassouli F, Tinschert P, Barata F, et al. Characteristics of Asthma-related Nocturnal Cough: A Potential New Digital Biomarker. J Asthma Allergy 2020;13:649-57. [Crossref] [PubMed]
11. Loudon RG, Brown LC. Cough frequency in patients with respiratory disease. Am Rev Respir Dis 1967;96:1137-43. [Crossref] [PubMed]
12. Loudon RG, Romans WE. Cough-monitoring equipment. Med Res Eng 1967;6:25-7.
13. Salmi T, Sovijärvi AR, Brander P, et al. Long-term recording and automatic analysis of cough using filtered acoustic signals and movements on static charge sensitive bed. Chest 1988;94:970-5. [Crossref] [PubMed]
14. Hall JI, Lozano M, Estrada-Petrocelli L, et al. The present and future of cough counting tools. J Thorac Dis 2020;12:5207-23. [Crossref] [PubMed]
15. Chaccour C, Sánchez-Olivieri I, Siegel S, et al. Validation and accuracy of the Hyfe cough monitoring system: a multicenter clinical study. Sci Rep 2025;15:880. [Crossref] [PubMed]
16. Kuhn M, Nalbant E, Kohlbrenner D, et al. Validation of a small cough detector. ERJ Open Res 2023;9:00279-2022. [Crossref] [PubMed]
17. Do W, Russell R, Wheeler C, et al. Performance of cough monitoring by Albus Home, a contactless and automated system for nocturnal respiratory monitoring at home. ERJ Open Res 2022;8:00265-2022. [Crossref] [PubMed]
18. Delmonico N. Cough and Wheeze Endpoint Measurement: Objective or Subjective? Applied Clinical Trials 2024. Accessed April 27, 2025. Available online: https://www.appliedclinicaltrialsonline.com/view/cough-and-wheeze-endpoint-measurement-objective-or-subjective-
19. U.S. Food and Drug Administration. 510(k) Premarket Notification: RESP Biosensor (K220893). U.S. Food and Drug Administration; 2022. Accessed April 27, 2025. Available online: https://www.accessdata.fda.gov/cdrh_docs/pdf22/K220893.pdf
20. Galvosas M, Gabaldón-Figueira JC, Keen EM, et al. Performance evaluation of the smartphone-based AI cough monitoring app - Hyfe Cough Tracker against solicited respiratory sounds. F1000Res 2022;11:730. [Crossref] [PubMed]
21. Hyfe. Overview of regulatory status for current cough counting devices on the market. Hyfe; 2023. Accessed April 27, 2025. Available online: https://www.hyfe.com/insights/overview-of-regulatory-status-for-current-cough-counting-devices-on-the-market
22. Puente-Maestu L, Dávila I, Quirce S, et al. Burden of refractory and unexplained chronic cough on patients' lives: a cohort study. ERJ Open Res 2023;9:00425-2023. [Crossref] [PubMed]
23. Dicpinigaitis PV, Morice AH, Smith JA, et al. Efficacy and Safety of Eliapixant in Refractory Chronic Cough: The Randomized, Placebo-Controlled Phase 2b PAGANINI Study. Lung 2023;201:255-66. [Crossref] [PubMed]
24. Abubakar-Waziri H, Edwards DA, Bhatta DB, et al. Inhaled alkaline hypertonic divalent salts reduce refractory chronic cough frequency. ERJ Open Res 2024;10:00241-2024. [Crossref] [PubMed]
25. Mazzone SB, Satia I, McGarvey L, et al. Chronic cough and cough hypersensitivity: from mechanistic insights to novel antitussives. Lancet Respir Med 2022;10:1113-5. [Crossref] [PubMed]
26. Choate R, Pasquale CB, Parada NA, et al. The Burden of Cough and Phlegm in People With COPD: A COPD Patient-Powered Research Network Study. Chronic Obstr Pulm Dis 2020;7:49-59. [Crossref] [PubMed]
27. Burgel PR, Nesme-Meyer P, Chanez P, et al. Cough and sputum production are associated with frequent exacerbations and hospitalizations in COPD subjects. Chest 2009;135:975-82. [Crossref] [PubMed]
28. Crooks MG, den Brinker AC, Thackray-Nocera S, et al. Domiciliary Cough Monitoring for the Prediction of COPD Exacerbations. Lung 2021;199:131-7. [Crossref] [PubMed]
29. Sumner H, Woodcock A, Kolsum U, et al. Predictors of objective cough frequency in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2013;187:943-9. [Crossref] [PubMed]
30. Griffith DE, Levin AR, Rudd M, et al. Using Continuous Cough Monitoring to Assess Bronchiectasis Therapy. Am J Respir Crit Care Med 2024;209:A7084.
31. Irwin RS, French CL, Chang AB, et al. Classification of Cough as a Symptom in Adults and Management Algorithms: CHEST Guideline and Expert Panel Report. Chest 2018;153:196-209. [Crossref] [PubMed]
32. Lopez V, Houana A, Sanchez-Olivieri I, et al. Clinical experience using the Hyfe cough tracker in malaria patients (abstract). Presented at: 71st Annual Meeting of the American Society of Tropical Medicine and Hygiene (ASTMH); Seattle, Washington; October 30–November 3, 2022.
33. Zimmer AJ, Ugarte-Gil C, Pathri R, et al. Making cough count in tuberculosis care. Commun Med (Lond) 2022;2:83. [Crossref] [PubMed]
34. Small P, Rudd M, Chung KF. Cough-omics: Quantifying the Variability and Predictability of Daily Cough Counts in Persistent Coughers. American Journal of Respiratory and Critical Care Medicine 2024;209:A7081.
35. Holt KJ, Smith JA. How often do healthy people cough? Respir Res 2023;24:275. [Crossref] [PubMed]
36. Hsu JY, Stone RA, Logan-Sinclair RB, et al. Coughing frequency in patients with persistent cough: assessment using a 24 hour ambulatory recorder. Eur Respir J 1994;7:1246-53. [Crossref] [PubMed]
37. Gabaldón-Figueira JC, Keen E, Rudd M, et al. Longitudinal passive cough monitoring and its implications for detecting changes in clinical status. ERJ Open Res 2022;8:00001-2022. [Crossref] [PubMed]
38. Hyfe. Whitepaper: The importance of prolonged cough monitoring: Data-driven evidence for a 7-day standard. Hyfe; 2024. Accessed April 27, 2025. Available online: https://www.hyfe.com/white-papers/the-importance-of-prolonged-cough-monitoring-data-driven-evidence-for-a-7-day-standard
39. Lee SE, Rudd M, Kim TH, et al. Feasibility and Utility of a Smartphone Application-Based Longitudinal Cough Monitoring in Chronic Cough Patients in a Real-World Setting. Lung 2023;201:555-64. [Crossref] [PubMed]
40. Holt KJ, Dockry RJ, McGuinness K, et al. An exploration of clinically meaningful definitions of cough bouts. ERJ Open Res 2024;10:00316-2024. [Crossref] [PubMed]
41. Ryerson CJ, Abbritti M, Ley B, et al. Cough predicts prognosis in idiopathic pulmonary fibrosis. Respirology 2011;16:969-75. [Crossref] [PubMed]
42. Wang Y, Zhang X, Chakalasiya JM, et al. HearCough: Enabling continuous cough event detection on edge computing hearables. Methods 2022;205:53-62. [Crossref] [PubMed]
43. Chamberlain Mitchell SA, Garrod R, Clark L, et al. Physiotherapy, and speech and language therapy intervention for patients with refractory chronic cough: a multicentre randomised control trial. Thorax 2017;72:129-36. [Crossref] [PubMed]
44. Slovarp LJ, Reynolds JE, Tolbert S, et al. Cough desensitization treatment for patients with refractory chronic cough: results of a second pilot randomized control trial. BMC Pulm Med 2023;23:148. [Crossref] [PubMed]
45. Sundholm N, Shelly S, Wright ML, et al. Effect of Behavioral Cough Suppression Therapy Delivered via Telehealth. J Voice 2025;39:851.e1-5. [Crossref] [PubMed]
46. Hyfe Inc. Hyfe to Launch World’s First Prescription Digital Therapeutic to Treat Chronic Cough [news release]. Published February 24, 2025. Accessed April 27, 2025. Available online: https://www.hyfe.com/news/hyfe-to-launch-worlds-first-prescription-digital-therapeutic-to-treat-chronic-cough
47. U.S. Food and Drug Administration. Regulatory Considerations for Prescription Drug Use-Related Software. U.S. Food and Drug Administration; 2023. Accessed April 27, 2025. Available online: https://www.fda.gov/regulatory-information/search-fda-guidance-documents/regulatory-considerations-prescription-drug-use-related-software
48. Gabaldón-Figueira JC, Keen E, Giménez G, et al. Acoustic surveillance of cough for detecting respiratory disease using artificial intelligence. ERJ Open Res 2022;8:00053-2022. [Crossref] [PubMed]
49. Schelfhout J, Nguyen AM, Birring SS, et al. Validation and Meaningful Change Thresholds for an Objective Cough Frequency Measurement in Chronic Cough. Lung 2022;200:717-24. [Crossref] [PubMed]