Understanding the economic burden of chronic cough: a systematic literature review

Abstract

Chronic cough (CC) is associated with high healthcare resource utilization (HCRU) due to challenges in diagnosis and treatment and is anticipated to have a substantial economic impact. This systematic literature review (SLR) sought to identify evidence on the cost-effectiveness of treatments and the economic burden associated with CC. Electronic database searches were supplemented with searches of conference proceedings and health technology assessment body websites. Two independent reviewers assessed all citations for inclusion based on predefined inclusion/exclusion criteria. Key inclusion criteria were patient population with CC, and outcomes related to cost-effectiveness and HCRU and costs. After screening, one cost-effectiveness analysis was identified, alongside eight studies reporting HCRU and costs related to CC. Though evidence was limited, studies suggest that patients with CC incur higher costs and use more resources than those with acute cough. Types of resource use reported included healthcare contacts and prescriptions, diagnostic tests, referrals and specialist evaluations, and treatment use. There is a paucity of literature on HCRU and costs in CC, and very limited cost-effectiveness analyses. The economic burden appears higher in these patients however, without direct comparison to the general population it is difficult to determine the total impact. The increased burden is expected to be a result of the challenges with diagnosis and lack of approved treatments. However, limited conclusions can be drawn in the absence of further data. Future studies should endeavor to quantify the HCRU and cost attributable to patients with CC.

Take home message

Patients with CC encounter multiple physicians and specialists to seek a definitive diagnosis, resulting in delayed diagnosis and appropriate treatment. This SLR reported a paucity of economic data; however, it found patients with CC to incur higher costs and more resource use than those with acute cough.

Background

Chronic cough (CC) is typically defined as a cough persisting for more than eight weeks [1,2,3,4]. Although cough is a common reflex mechanism, excessive and prolonged cough can be highly burdensome to patients, disrupting daily activities for the individual and those around them [2, 4, 5]. In severe cases, patients with CC may experience complications such as stress urinary incontinence, interference with speech, depression, and syncope, which can have a substantial negative impact on quality of life (QoL) and on activities of daily living [4, 6].

Many patients with CC are believed to have cough reflex hypersensitivity, in which the reflex response occurs in response to low levels of stimulation from thermal, chemical or mechanical sources [4]. In some cases, CC is associated with an underlying comorbid condition, such as asthma, gastroesophageal reflux disease (GERD), or upper airway cough syndrome (UACS). When cough persists despite treatment of the associated condition, this is defined as refractory chronic cough (RCC) [6]. Conversely, unexplained chronic cough (UCC) may be diagnosed when all other aetiologies have been excluded through a thorough clinical evaluation and work-up [7]. In both RCC and UCC, cough reflex hypersensitivity has been proposed to contribute to the pathophysiology of CC [6, 8]. This review focuses on CC, inclusive of RCC and UCC.

CC has been estimated to affect around 10% of the adult population [5, 9]. A pooled analysis estimated an overall prevalence of 9.6%, with higher regional prevalence rates in Oceania (18.1%), Europe (12.7%), and America (11.0%), than in Asia (4.4%) and Africa (2.3%) [5]. Overall, population-based prevalence estimates of CC are limited [9].

Patients presenting with CC progress through evidence-based assessments to evaluate and diagnose treatable traits of the disease in attempts to offer direct therapy [4]. Patients often interact with multiple physicians and receive several specialist referrals in an attempt to seek a definitive diagnosis [10,11,12]. Until 2021, there was no CC specific diagnosis code [13]. Delays to diagnosis and appropriate treatment may therefore occur as different therapeutic options are tried [6]. The European Respiratory Society (ERS) guidelines recommend sequential therapeutic trials of a number of agents in turn, with treatment ceased if no responses are observed [4]. Currently trial therapies include antitussives, protussives, inhaled corticosteroids (ICS), bronchodilators, and neuromodulators, such as opioids, pregabalin, and gabapentin; however, none of these are currently approved for use in CC [4]. Indeed, at the time of writing, there are no approved pharmacological treatments for CC, however, a number of clinical trials investigating novel treatments for CC are underway [4].

Given the increased healthcare resource utilization (HCRU) required to determine a diagnosis of CC and trial of different therapeutic interventions, CC is expected to have a substantial economic impact. This may include both direct costs such as physician costs, medication costs, diagnosis costs and hospitalization, as well indirect costs resulting from productivity losses and absenteeism from work. There is currently limited evidence on the economic burden of CC, and in particular the burden specifically attributable to RCC and UCC. The lack of approved therapies for CC may contribute to the paucity of economic evaluations in this therapy area. A systematic literature review (SLR) was performed to identify evidence on the cost-effectiveness of treatments and the economic burden associated with respect to HCRU and costs attributable to CC.

Methods

Search strategy

A search was performed on 25 February 2021 in the MEDLINE, EMBASE, EconLit and EBM Reviews electronic databases via the Ovid platform. A combination of free text searching, and subject headings were used to capture the target population and outcomes. Study design filters for economic evidence recommended by the Scottish intercollegiate Guidelines Network (SIGN) were used, [14] supplemented with terms from alternative search strategies recommended by the InterTASC Information Specialists’ Sub-Group Search Filter Resource to increase the sensitivity of the search [15]. The complete search strings are presented in Supplementary Tables 1, Supplementary Tables 2, Supplementary Tables 3, and Supplementary Table 4.

The NHS Economic Evaluation Database, Health Economic Evaluations Database, and Tufts Cost-Effectiveness Analysis Registry were hand-searched using key population search terms to identify relevant studies. Hand searches of conference proceedings of annual meetings of relevant societies from the two years prior to search date were used to augment the database searches. Societies included the American Academy of Allergy, Asthma, and Immunology, American College of Allergy, Asthma, and Immunology, American College of Chest Physicians, American Thoracic Society, and European Respiratory Society. Health technology assessments (HTA) evaluating therapies for CC as published by the National Institute of Health and Care Excellence (NICE), Canadian Agency for Drugs and Technologies in Health (CADTH), Agency for Healthcare Research and Quality (AHRQ), Institute for Clinical and Economic Review (ICER), National Institute for Health and Care Research Health Technology Assessment (NIHRHTA), International Network of Agencies for Health Technology Assessment (INAHTA), Haute Autorité de Santé (HAS) and Institute for Quality and Efficiency in Health Care (IQWiG) were also screened.

The same search strategy was used to identify cost-effectiveness and HCRU studies, as current study design filters do not reliably discriminate between these types of studies [15].

Study selection

Studies were assessed for inclusion based on the PICOTS criteria outlined in Table 1. The target population was adult patients diagnosed with RCC or UCC, according to American College of Chest Physicians (ACCP) guidelines [16]. Due to the heterogeneity in defining CC across studies, the population search strings were expanded to include all patients with CC as defined by the study investigators. Cost-effectiveness and HCRU studies were assessed against the same eligibility criteria, with the exception of the intervention and comparators. Cost-effectiveness studies were considered for inclusion if any medication known to be used for the treatment of CC, including off-label medications, compared to placebo, best supportive care, or any other intervention of interest, were reported. HCRU studies were not restricted by intervention or comparator to account for studies reporting costs and/or resource use independent of treatment effects. Outcomes of interest included costs combined with measures of effectiveness, and HCRU outcomes such as total healthcare costs, direct costs, indirect costs, out-of-pocket costs, and resource utilization. Relevant studies were limited to English language publications only, and no time or geographical restrictions were imposed.

Table 1 Eligibility criteria for SLR study inclusion

Screening of all titles and abstracts identified in the search was conducted by two independent reviewers. Citations considered eligible for inclusion by both reviewers were advanced to full-text screening, which involved independent assessment of the full-text articles for inclusion by the same two reviewers. A third reviewer provided arbitration in the case of discrepancy. Each study was counted once through mapping of citations to corresponding studies.

Data extraction and quality assessment

Data extraction from the included citations was undertaken by two independent reviewers, with a third reviewer to reach consensus for any discrepancies remaining following reconciliation. Extracted data included study identifiers, study characteristics, intervention characteristics, patient characteristics, and outcomes.

One reviewer assessed the quality of included studies, with judgments validated by a senior reviewer, using the Drummond checklist [17].

Results

Identification of studies

Cost-effectiveness

A total of 1742 cost-effectiveness citations were identified through electronic databases. Supplementary searches of conference proceedings and HTA websites identified a further six citations. After removal of 257 duplicates, the titles and abstracts of 1491 unique citations were screened, of which 18 were retrieved for full-text review. The majority of citations excluded at the abstract screening stage (829/1473) were excluded based on population. Only one of the 18 studies advanced to full-text review was determined to meet the inclusion criteria. Of those excluded, nine were excluded based on outcome, five based on population, and three based on study design. Due to limited evidence, data were unable to be stratified based on subgroups of interest. A PRISMA flow diagram for the study selection process is presented in Fig. 1.

Fig. 1

PRISMA flow diagram for cost-effectiveness studies

HCRU and costs

Both HCRU and cost-effectiveness searches were undertaken using the same search strategy, therefore the same number of citations were identified through electronic database searches (n = 1742) and supplementary searches (n = 6) for HCRU. The titles and abstracts of 1491 unique citations were screened, of which 18 were retrieved for full-text review. The majority of citations excluded at the abstract screening stage (829/1473) were excluded based on population. Following full-text review, ten records were excluded, yielding a total of eight unique studies reporting HCRU and costs. No studies focusing on costs or HCRU in patients with a CC diagnosis of ≥ 1 year or < 1 year were identified. Of those excluded, five were excluded based on population, three based on study design, and two based on outcome. A PRISMA flow diagram for the study selection process is presented in Fig. 2.

Fig. 2

PRISMA flow diagram for healthcare resource utilization studies

An overview of the included studies is presented in Table 2.

Table 2 Identified studies reporting cost-effectiveness, HCRU and/or costs in CC

Cost-effectiveness analyses

The single cost-effectiveness analysis identified for CC was conducted in Singapore and published as a journal article in 2001 [18]. The study involved a decision tree model to assess the cost-effectiveness of six alternative management strategies for UCC lasting at least three weeks: [1] Test all then treat; [2] Treat all; [3] Treat postnasal drip syndrome (PNDS), test asthma, treat gastroesophageal reflux disease (GERD); [4] Treat sequentially starting with PNDS; [5] Test then treat sequentially; and [6] Treat PNDS, test asthma and GERD together. Treat all, Test then treat sequentially, and Treat sequentially starting with PNDS were found to be the most cost-effective strategies considering the expected duration of cough versus expected cost, with direct costs of $157, $149, and $184 USD, respectively [18].

Costs associated with CC

There is a paucity of cost data in CC, with only four studies reporting costs, as outlined in Table 3. Of these, two were conducted in the United Kingdom (UK), [11, 19] one in the United States (US), [10] and one in Singapore [18]. Two studies were published as journal articles, whilst two were presented as conference abstracts. Of note, a retrospective study comparing CC and acute cough in the UK found that patients with CC incurred higher costs than those with acute cough (all healthcare contact, per person-year equaled £3,663 in patients with CC and £2,700 in acute cough) [19]. The impact of redesigning the cough clinical care process in the US was assessed and it was proposed that by introducing a multidisciplinary, collaborative approach to care, the costs could be nearly halved ($656 vs. £$1,319 USD in current usual care) [10]. No studies explored indirect costs associated with CC.

Table 3 Summary of studies reporting costs associated with CC

HCRU associated with CC

Resource use was explored in seven studies, as outlined in Table 4. HCRU in the US was evaluated in five studies, [10, 12, 20,21,22] whilst one study assessed HCRU in the UK, [19] and one in Singapore [18]. Only two studies were published as journal articles, whilst the remaining five were conference abstracts. Types of resource use reported included healthcare contacts and prescriptions, diagnostic tests, referrals and specialist evaluations, and treatment use. Differences in the resource use reported across studies prevent cross-country comparisons, however it is clear that resource use is high in patients with CC. A retrospective study in the US compared patients with and without CC and found an increase in prescription of opioid-containing cough suppressants (OCCS) in patients with CC (60 prescriptions per 100-patients in CC cohort vs. 12 prescription per 100-patients in non-CC cohort) [21]. 35% of patients with CC in this study reported ≥ 3 OCCS prescriptions, with 0.6% reporting ≥ 10 OCCS prescriptions [21]. Data collected from a retrospective database study identified chest x-rays (80.3%), systemic respiratory antibiotics (72.4%) and narcotics, including codeine (60.9%) as the most common causes for healthcare interactions [12].

Table 4 Summary of studies reporting HCRU associated with CC

Study quality assessment

The quality of the eight studies included in the two SLRs was assessed using the Drummond checklist in terms of their clarity of reporting in ten different areas: study question, selection of alternatives, form of evaluation, effectiveness data, benefit measurement, costing, modelling, analysis and interpretation of results, allowance for uncertainty, presentation of results [17]. Overall, the quality of the included studies was moderate to low, often because the short length of conference abstracts did not allow for complete reporting.

Conclusions

There is a paucity of literature on HCRU and costs in CC, and very limited cost-effectiveness analyses in this population possibly due to lack of approved therapies for CC. The findings of this SLR indicate that the economic burden appears to be higher in patients with CC however, without direct comparison to the general population is it difficult to conclude on the level of impact. The increased burden may be expected due to challenges with diagnosis which may result in multiple physician visits and referrals, an increased number of comorbidities in patients with CC compared to patients without CC, and lack of approved treatments. It is likely that any current estimates of the burden in patients is underestimated due to widespread underdiagnosis. However, limited conclusions can be drawn in the absence of further data. Future studies involving cough registries might help to better quantify the HCRU and costs attributable to patients with CC.

Discussion

CC is a highly burdensome condition affecting approximately 10% of the adult population [5]. Symptoms of CC include chest pains, sleep disturbance and hoarse voice, and in severe cases, syncope, stress incontinence and vomiting [3]. Despite the high burden CC places on patients, diagnosis of CC remains a challenge, causing significant delays in treatment [6, 16]. As a result, CC is expected to have a high economic burden including both direct costs such as costs for physician visits, medication, diagnosis and hospitalization, and indirect costs including loss of productivity and absenteeism from work. This SLR was performed to identify and describe current literature on the economic burden of RCC and UCC, including the relative cost-effectiveness of current treatments.

The SLR revealed that there is a lack of evidence reporting the economic burden of CC, with notable gaps in the reporting of out-of-pocket expenses, costs for over-the-counter medication, and the costs associated with discrete treatments. Overall, only six conference abstracts and two journal articles were identified reporting on the economic impact of CC. Only one study focused on UCC, [18] and no studies evaluated RCC. Although one cost-effectiveness analysis was identified, no studies reported incremental costs, incremental life years, or incremental QALYs [18]. Notably, the majority of citations were excluded based on population, further emphasizing the lack of evidence in this disease area, and the heterogeneity in defining CC. Whilst the limited data on the economic impact of CC may reflect difficulties in collecting and analyzing such data, for example due to diagnostic challenges, there is a clear need for future studies to focus on quantifying this burden. Following the completion of this SLR, the ICD-10 Coordination and Maintenance Committee in the US implemented a new ICD-10-CM code in October 2021 specific for the diagnosis of CC (R05.3) [13] This addition may address some of the difficulty in identifying the economic burden of patients with CC however, until there is consistent implementation and diagnoses with a standardized code there will remain an uncertainty in the true economic burden of CC.

Despite the limited evidence base, collectively it is shown that patients with CC use more healthcare resources and incur greater costs than patients with acute or non-CC [19]. Although most patients with CC are managed in a primary care setting, up to half of patients are referred to one or more specialists [12, 20, 22]. When compared to patients without CC, a recent population-based study confirmed that patients with CC encountered more (1.5 times) visits to health care providers compared to a non-CC control group (6.7 vs. 4.4; p = < 0.001) [9]. Patients with CC receive an average of 2.1 diagnostic tests, with most undergoing chest x-ray and many undergoing advanced chest imaging [20]. As many as half of patients with CC are treated with opiates or other narcotics, which is of pressing concern given that overreliance on opioid prescriptions is a driving factor of the current opioid crisis [12, 21, 23]. From a financial perspective, the average direct cost incurred by patients with CC was reported to be £1,800 GBP over a 12 month period in the UK and $1,319 in the US [10, 11]. However, these data are based on individual studies and so should be taken with caution. Calculating cost-effectiveness in CC using traditional measures (i.e., number of hospitalizations) poses its own difficulty as patients with CC are typically not hospitalized due to CC itself, this is often attributed to a comorbid condition. No studies reported on the indirect costs incurred. Further, the broad definition of CC and heterogeneity among patients, compounded by the high prevalence of comorbidities, causes challenges in estimating costs directly attributable to CC. It is difficult to compare these costs to the general population as no studies examined the costs for patients with CC in the context of the general population.

Strengths and limitations

The SLR involved highly sensitive database searches of peer-reviewed literature as well as searches of recent conferences and HTA body websites to identify unpublished studies. Validated search filters recommended by SIGN and InterTASC were used to prioritize sensitivity over specificity of the searches [14, 15]. However, as with any SLR, there is a risk that studies published after the database searches conducted in February 2021 may not have been captured.

A particular weakness of this SLR is that the limited amount of available published data prevents comprehensive or definitive conclusions from being drawn on the economic impact of CC. The variation in follow-up duration, ranging from weeks to years, prevents an accurate characterization of costs and healthcare resource utilization. The bulk of the evidence included in this SLR came from conference abstracts, which do not provide complete information and should be interpreted with caution as they do not undergo the same methodologically rigorous peer review process as fully published results. Further, CC involves long-term management, and there is a clear lack of studies to adequately assess the true long-term costs of the condition.

Of note, there are difficulties in identifying relevant data and performing appropriate analysis, given the broad definition of CC and challenges with diagnosis. In this SLR, the target population was adult patients diagnosed with CC, either refractory or unexplained, according to ACCP guidelines [16]. However, broader inclusion criteria were applied to ensure capture of all relevant data. Studies evaluating any type of CC as defined by the study investigators regardless of its duration, including idiopathic CC, as well as studies using alternative definitions of RCC or UCC, were also included. Further, it can be difficult to isolate the burden specifically associated with CC due to the high prevalence of comorbidities in these patients. In previous studies it was noted patients with CC were more likely to be smokers and/or obese, and to have respiratory or airway diseases, psychological disorders, diabetes, or chronic pain [24]. These comorbidities are likely to attribute to greater medical attention and costs.

Overall, whilst evidence is limited, CC is associated with high utilization of healthcare resources involving multiple patient referrals, diagnostic tests, and drug prescriptions. Combined with the delayed diagnosis and limited treatment options in this population, there is high humanistic and economic burden, with a remaining unmet need for a more effective treatment approach to reduce the associated burden.

Take home message

Patients with CC encounter multiple physicians and specialists to seek a definitive diagnosis, resulting in delayed diagnosis and appropriate treatment. This SLR reported a paucity of economic data; however, it found patients with CC to incur higher costs and more resource use than those with acute cough.