NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

McCrory DC, Coeytaux RR, Yancy WS Jr, et al. Assessment and Management of Chronic Cough [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2013 Jan. (Comparative Effectiveness Reviews, No. 100.)

Cover of Assessment and Management of Chronic Cough

Assessment and Management of Chronic Cough [Internet].

Show details

Results

Introduction

In what follows, we begin by describing the results of our literature searches. We then provide a brief description of the included studies. The remainder of the chapter is organized by Key Question (KQ). Under each of the two KQs, we begin by listing the key points of the findings, followed by a brief description of included studies, followed by a more detailed synthesis of the evidence. The detailed syntheses under KQ 1 are organized by measures of cough frequency, cough severity, and responsiveness. The detailed syntheses under KQ 2 are organized by comparison drug classes (antitussives, protussives, other agents). We conducted quantitative syntheses where possible, as described in the Methods chapter.

A list of abbreviations and acronyms used in this chapter is provided at the end of the report.

Results of Literature Searches

Figure 2 depicts the flow of articles through the literature search and screening process. Searches of PubMed®, Embase®, and CDSR yielded 21,860 citations, 6,504 of which were duplicate citations. Manual searching identified 75 additional citations, for a total of 15,431 citations. After applying inclusion/exclusion criteria at the title-and-abstract level, 833 full-text articles were retrieved and screened. Of these, 718 were excluded at the full-text screening stage, leaving 115 articles for data abstraction. These 115 articles described 121 unique studies, 78 of which were relevant to KQ 1, and 48 of which were relevant to KQ 2 (5 studies were relevant to both KQs). No additional information was found through our grey literature search.

This figure depicts the flow of articles through the literature search and screening process. Searches of PubMed, Embase, and the Cochrane Database of Systematic Reviews yielded 21,860 citations, 6504 of which were duplicate citations. Manual searching identified 75 additional citations, for a total of 15,431 citations. After applying inclusion/exclusion criteria at the title-and-abstract level, 833 full-text articles were retrieved and screened. Of these, 718 were excluded at the full-text screening stage, leaving 115 articles (representing 121 unique studies) for data abstraction. Total numbers of studies included for each KQ were 78 for KQ 1, and 48 for KQ 2.

Figure 2

Literature flow diagram. KQ=Key Question; RCT=randomized controlled trial a Five studies were relevant to both KQ 1 and KQ 2.

Appendix C provides a detailed listing of included articles. Appendix D provides a complete list of articles excluded at the full-text screening stage, with reasons for exclusion.

Description of Included Studies

Overall, we included 121 studies represented by 115 publications: 78 studies were relevant to KQ 1, 48 to KQ 2. Studies were conducted in Europe (54%); the United States or Canada (23%); Australia or New Zealand (11%); Asia (8%); and other locations (8%). Nineteen studies in KQ 1 (23%) and 3 studies in KQ 2 (6%) included children. Forty-five studies (37%) were published before 2000.

Further details are provided in the relevant KQ results sections, below.

Key Question 1. Instruments Used To Assess Cough

KQ 1: In adults and adolescents (≥ 14 years of age) and children (<14 years of age), what is the comparative diagnostic accuracy, therapeutic efficacy, and patient outcome efficacy of instruments used to assess cough?

Key Points

  • Electronic recording devices are accurate for assessing cough frequency, but they show variable correlation with instruments that measure other dimensions of cough.
  • The Leicester Cough Questionnaire (LCQ) and the Cough-specific Quality of Life Questionnaire (CQLQ) are the most widely studied cough-specific quality-of-life questionnaires in adult populations. Both have demonstrated validity and reliability, with emerging evidence available on responsiveness.
  • There is moderate strength of evidence to support the validity and responsiveness of the Parent Cough-specific Quality of Life Questionnaire (PC-QOL) in assessing the severity/QOL of cough among children.
  • Emerging data support the responsiveness of recording devices, cough-related questionnaires, and tussigenic challenge tests, but further research is needed to accurately estimate the minimally important difference (MID) of these assessment instruments.
  • Although diaries and visual analog scales are based on face validity, assess a wide variety of different cough outcomes, and are widely used both in research and practice, there are few data to validate their accuracy in assessing cough, and what data exist show inconsistent correlations with other cough measurement tools. These tools are usually simple and easy to use, but more data are needed to determine their reliability and validity in assessing cough frequency or severity/QOL.
  • While all of the included studies evaluated aspects of the comparative diagnostic accuracy of the various cough measurement tools, none evaluated the comparative therapeutic efficacy or patient outcome efficacy of these tools.

Description of Included Studies

Cough can be assessed along several dimensions, the most of important of which may be frequency, severity, and cough-specific QOL. Cough frequency is objective and relatively easy to measure but may not necessarily correlate with severity or cough-specific QOL, whereas cough severity and cough-specific QOL may be closely interrelated. Most of the standardized questionnaires included in this report measured aspects of both of these latter dimensions. Therefore, for the purpose of this report, we considered instruments that measured both severity and QOL together to be “severity/QOL” instruments. In this CER we evaluate the available data that support the validity and reliability of instruments to measure one of two dimensions of cough: (1) cough frequency; or (2) the severity/QOL impact of cough (including assessments of the impact of cough on sleep, work, general well-being, health-related quality of life, etc.). We also evaluate the available data that support these instruments' ability to measure potentially meaningful clinical change over time (responsiveness).

To be eligible for inclusion in this report, a study had to either: 1) compare a cough frequency or severity/QOL assessment instrument with one or more cough assessment, health-related quality of life, or clinical change instrument; or 2) report data on changes in the instrument score over time in response to treatment for cough or the underlying etiology of the cough. For the purposes of this report, we consider tussigenic challenge tests and exhaled nitric oxide tests as severity/QOL assessments.

In what follows, we rely heavily on tabular presentation of information because of the large degree of heterogeneity with respect to patient populations, study design and objectives, index and reference tests, and the nature and content of the results reported in the individual studies. We summarize the findings of studies of adults and adolescents (≥14 years of age) separately from those of children (<14 years of age). Studies that include adults, adolescents, and children are listed only once in a given table (categorized as “Studies in Adults, Adolescents, and Children”). These studies are included in the study counts in both the “adult and adolescents” and the “children” sections in the text below. Note, however, that their findings are summarized in one or the other of these sections based on the mean age of the included patients and therefore the relevance of the findings to the overall adult or pediatric populations.

A total of 78 studies met the inclusion criteria for this KQ.8,31-106 Seven were RCTs;34,56,59,80,93,96,100 the remaining 71 were observational studies. Nineteen studies (24%) were conducted in the United States or Canada,32,43,45,47,48,50,56,58,60,63,67,74,79,88,90,94,99,102,103 and 39 (50%) in Europe,34,36-39,44,49,51-53,55,59,61,62,64,66,68,70-73,75,77,78,80,81,83,87,89,91,92,95-98,101,105-107 including 32 (41%) conducted exclusively in the UK.36-38,44,49,51-53,55,59,61,62,64,68,70-73,77,78,80,81,83,89,91,92,95-98,101,107 Seven studies took place in Australia/New Zealand,40-42,84-86,104 five in other locations,57,69,76,82,93 and four in multiple locations.46,54,65,100 Geographical location was not reported for four studies.8,31,33,35 In most cases, the funding source was not reported or was unclear. Other study characteristics are summarized in Appendix F (Table F-1). A total of 5,927 participants were included across studies; sample sizes of individual studies ranged from 1 to 671 subjects. Of the 78 studies, 67 (86%) were judged to have a low risk of bias, and 11 (14%) were judged to have a high risk of bias (see Appendix E for details).

Thirty-three studies (42%) enrolled patients with chronic cough of mixed, unknown, or unspecified etiology; 18 (23%) enrolled patients with acute cough or cough of unspecified duration, and 27 (35%) focused on specific clinical conditions such as chronic bronchitis, asthma, or lung cancer. Fifty-nine studies included adults and adolescents (≥14 years of age), 15 included only children (<14 years of age), and 4 included adults, adolescents, and children.

Thirty-two studies (41%) identified a specific cough-related questionnaire as an index test, 22 studies (28%) reported on a specific electronic device designed to measure cough frequency, and 24 (31%) reported on instruments other than cough-related questionnaires or electronic recording devices. Most studies that included other instruments such as visual analog scales, symptom diaries, tussigenic challenges, or self-reported global change items utilized them as validation tools but did not evaluate them as index tests themselves. Few of the included studies provided information regarding previous validation of reference tests. While all of the included studies evaluated aspects of the comparative diagnostic accuracy of these measurement tools, none evaluated their comparative therapeutic efficacy or patient outcome efficacy.

Detailed Synthesis

Measures of Cough Frequency

Our search identified 42 studies that evaluated instruments designed to assess the frequency of cough (Table 5).35-37,41,44,45,47,55,59,62,64,65,71,73-75,77,78,90,95,98,102 Of the 42 studies, 8 (19%) were conducted in the United States or Canada.45-47,74,90,94,102,103 Thirty-seven studies (88%) were judged to have low risk of bias, and 5 (12%) had high risk of bias. A variety of reference standards were employed to validate these instruments, including human count, other electronic recording devices, video recording devices, quality-of-life questionnaires, subjective scoring, and laboratory tussigenic challenges.

Table 5. Results of studies of cough frequency assessment instruments.

Table 5

Results of studies of cough frequency assessment instruments.

Twenty-four different cough frequency assessment tools were described in the 42 studies summarized above (Appendix F, Table F-2), not including humans counting coughs either during direct observation of patients or from recording devices. Of these 24 cough frequency assessments, all but one (the Fisman Cough Severity Frequency Score) are electronic recording devices.

Adults and Adolescents (≥14 Years of Age)

Of the 42 studies that evaluated instruments designed to assess the frequency of cough, 29 (69%) included adults and adolescents, and 2 (5%) included adults, adolescents, and children (Table 5). Most of these studies evaluated the performance of electronic recording devices for the purpose of counting the number of coughs in a given period of time.

Five studies comparing sound recording devices with human cough count or video recording34,36,65,90,98 and one study comparing sound recording devices with another electronic recording device62 showed strong correlation between the measures. Note that the studies by Paul and colleagues90 and Hamutcu and colleagues62 also included children under 14 in their studies, although the mean age of the patients was 28.1 ± 25 and 13.6 ± 2.6 years, respectively. One study comparing sound recording devices with the LCQ showed moderately strong correlation.73 Other studies comparing sound recording devices with other cough measurement tools demonstrated only fair to moderate correlation.37,55,59,70,74,75,77,95,97 In general, the results of studies that evaluated electronic recording devices demonstrated variable sensitivity, good PPV, high correlation coefficients, and excellent specificity and NPV when comparing the recording devices with human cough counts.34-36,45,78

Three studies estimated the sensitivity and specificity of three different cough recording devices to assess cough frequency, with human cough counting as the reference standard.35,36,78 A fourth study102 calculated a correlation coefficient between cough counts as measured by an electronic recording device and human cough count. In all four studies, recording devices and humans reported nearly identical counts. This suggests that recording devices are highly valid as cough-counting instruments, at least in controlled or laboratory settings. Correlation between recording devices and other cough assessment instruments, however, was generally poor to moderate, with reported Spearman coefficient values generally in the 0.30–0.60 range. These findings are consistent with an interpretation of limited validity of recording devices; it is also possible, however, that counting coughs with an electronic recording is a valid way to assess cough frequency, but that cough frequency correlates only moderately with cough severity or QOL, which are the constructs that were usually assessed by the instruments with which recording devices were compared.

Five studies assessed the reliability of recording devices.35-37,55,102 Intraclass correlation coefficients were consistently ≥0.80, with reproducibility reported to be 100 percent in one study.35

Children (<14 Years of Age)

Of the 42 studies that evaluated instruments designed to assess the frequency of cough, 11 (26%) included only children under the age of 14, and 2 (5%) included adults, adolescents, and children (Table 5). All 13 of these studies reported on an electronic recording device. Reference tests included another electronic recording device (audio or video) in four studies,41,44,62,90 parent-reported questionnaires, scores, or diaries in five studies,40,42,61,64,101 child-reported scores or diaries in three studies,40,42,47 diaries where it was uncertain whether the parent or child was doing the reporting,31,53 and the PC-QOL instrument in one study.86 There was strong correlation (r ≥ 0.96) between electronic devices, and variable correlation between self-reported or parent-reported instruments and electronic recording devices. Some studies reported no significant relationship between parent or child reporting of cough frequency and the number of coughs identified by recording devices, whereas some reported a significant correlation during daytime but not during nighttime. One study64 reported a significant correlation (r=0.588, p<0.02) between cough frequency assessed by a cough recording device and a parental night cough diary during a week when medication for cough was administered to subjects, but no significant correlation during the placebo week.

Two studies estimated the sensitivity and specificity of two different cough recording devices to assess cough frequency, with human cough counting as the reference standard.44,47 Both demonstrated a sensitivity of approximately 80 percent.

Measures of Cough Severity or Quality of Life

We identified 56 studies that reported on instruments designed or purported to assess the severity of cough or the quality of life impact of cough on sleep, work, or an aspect of health-related quality of life (Table 6).8,31-34,37-40,42,43,46,48-52,54-61,63,66-69,71-73,76,77,79-86,88,91-97,99,100,104,106 A variety of reference standards were employed to validate these instruments, including electronic recording devices, quality-of-life questionnaires, subjective scoring, and laboratory tussigenic challenges. Of these studies, 14 (25%) were conducted in the United States or Canada.32,43,46,48,50,56,58,60,63,67,79,88,94,99,107 Forty-seven studies (84%) were judged to have a low risk of bias, and the remaining 9 (16%) to have high risk of bias.

Table 6. Results of studies of cough severity or quality-of-life impact assessments.

Table 6

Results of studies of cough severity or quality-of-life impact assessments.

Of the many cough frequency assessment tools described in the 56 studies summarized above, 14 were cough-related questionnaires (Table 7). Some of these are cough-specific (Leicester Cough Questionnaire [LCQ], Chronic Cough Impact Questionnaire [CCIQ], Cough-specific Quality of Life Questionnaire [CQLQ], Pediatric Cough Questionnaire [PCQ], Parent Cough-specific Quality-of-Life Questionnaire [PC-QOL], Adverse Cough Outcome Survey [ACOS]), while others focus on disease states for which cough is a predominant symptom (Chronic Bronchitis Symptoms Assessment Scale [CBSAS], Cough and Sputum Assessment Questionnaire [CASA-Q], Lung Cancer Cough Questionnaire [LCCQ], Punum Ladder, and Cough Severity Diary [CSD], and the Questionnaire for Lung Transplant Patients [QLTP]). Other instruments are general assessments, such as the Global Rating of Change (GRC) scale. The ACOS has been revised and replaced by the CQLQ. Twelve of the 14 questionnaires were developed and studied in adult populations, and 2 (the PCQ and PC-QOL) were designed for use in pediatric populations.

Table 7. Description of cough severity or quality-of-life impact instruments.

Table 7

Description of cough severity or quality-of-life impact instruments.

Adults and Adolescents (≥14 Years of Age)

Forty-five (80%) of the 56 studies that reported on instruments to assess the severity or quality of life impact of cough included adults and adolescents, and 2 (4%) included adults, adolescents, and children (Table 6). Note that the two studies which included adults, adolescents, and children63,69 had populations with a mean age of 38.3 and 6.8 years, respectively; therefore, the latter study by Hartnick and colleagues63 is discussed in more detail in the children only section, below, while the study by Kalpaklioglu69 is included here.

Of the cough-specific questionnaires, the LCQ was the most widely studied, with 19 studies. Four studies demonstrated strong correlation between the LCQ and other cough measurement tools such as electronic recording devices, subjective symptom scores, cough frequency scores, other questionnaires, and tussigenic challenges,38,68,73,81 while most of the remaining studies showed only fair to moderate correlation with a variety of other cough measurement tools, both objective and subjective.8,37,38,49,66,69,76,77,107 The LCQ was developed based on an outpatient adult population with chronic cough, and no studies evaluated its measurement accuracy in the pediatric population. The LCQ has the advantage of ease of administration and interpretation, which is ideal for an ambulatory clinic setting.

Data on the CQLQ were reported in 7 of the 55 studies on cough severity or quality of life impact.56,58,60,69,91,94,95 An additional three studies8,67,68 reported on the ACOS, which represents a prior generation of the CQLQ. The CQLQ, which includes six domains, has been shown to correlate with the LCQ, both of which appear to be better at assessing the impact of chronic cough than the SF-36.69 The CQLQ offers an advantage over the LCQ in its ability to capture sex differences in chronic cough because the LCQ, unlike the CQLQ, does not have an item that assesses urinary incontinence as an important side effect of cough.

The other cough-specific questionnaires have been less extensively studied, and although most report good internal correlation, results correlating these instruments with other cough measures or assessing responsiveness have been variable or lacking.

There is no universally accepted reference standard for the assessment of either cough severity or the impact of cough on health-related quality of life. Most studies of disease-specific or general quality-of-life questionnaires evaluate an instrument's validity by correlation of total scores or domain subscores with other cough or respiratory symptom measurement tools. Reliability of questionnaires in adult populations with cough was most commonly assessed by test-retest correlation and measures of internal consistency. Cronbach's alpha was generally high, with values > 0.80 reported for the majority of questionnaires. Repeatability was also generally good, with high intraclass coefficients reported for most of the questionnaires. In the absence of a single reference standard, however, and with application among a wide variety of patient populations, Spearman coefficients in the 0.2–0.8 range do not necessarily suggest that a given instrument is not a valid assessment tool. Nor is the Spearman coefficient an appropriate point of comparison between two different instruments. These findings may best be interpreted as providing evidence that some questionnaires are valid assessments of cough severity or the impact of cough in health-related quality of life, but that there is insufficient evidence to precisely characterize the validity of most of these instruments.

Children (p<14 Years of Age)

Nine of the 56 studies (16%) that reported on instruments to assess the severity or quality of life impact of cough included only children under the age of 14, and 2 (4%) included adults, adolescents, and children (Table 6). The vast majority of subjects in the one of these latter two studies were under the age of 14 (mean age was 6.8 years);63 we therefore consider this study to be primarily among children <14 years of age and discuss it below.

Only two named cough-related questionnaires (PC-QOL and PCQ) were evaluated by these 8 studies. In a series of three studies,84-86 Newcomb and colleagues compared the PC-QOL and the PedsQL (a generic quality-of-life instrument), a cough score, the SF-36, VAS, and VCD instruments, and cough counts. Variable correlation was demonstrated with domains of the PedsQL and SF-36. Robust correlation was demonstrated with VAS and VCD scores (r=-0.55 and r=-0.70, p<0.001) but not with cough counts (r=-0.32, p=0.074). Chang et al.104 administered the PC-QOL, the PedsQL (a generic quality-of-life instrument), and a cough diary to 346 children to assess the burden and etiologies associated with chronic cough, and found that the PC-QOL did not correlate with duration of cough (r=-0.01, p=0.92).

A single study evaluated the PCQ. In this study,63 Hartnick et al. performed validation exercises by administering the 5-item PCQ to the parents of 120 children with chronic cough on 3 different occasions. The first two administrations were conducted within 2 weeks of each other, prior to initiating treatment. The third administration was conducted 3 weeks after the second one to determine if the PCQ would accurately reflect parents' perception of how their child's cough had changed following treatment, as assessed by a parent-reported global assessment of change. This study demonstrated test-retest reliability for each of the five PCQ questions, and provided evidence to confirm the PCQ's internal consistency, discriminant validity, and convergent validity.

Three other studies conducted in pediatric populations did not evaluate a named questionnaire. Two compared diaries or cough severity/QOL scores completed by parents and children to an electronic cough monitor,40,42 and one compared a parent questionnaire and parent diary with video recordings and an assessment of children's sleep.61 The results of these three studies suggest that the frequency of children's cough is variably related to parents' self-reported assessments, with a generally stronger relationship between cough frequency and parents' assessment of cough severity or QOL impact during the daytime than at nighttime.

Instrument Responsiveness

Of the 78 studies pertinent to KQ 1 identified by our literature search, 36 (46%) provided information on at least one instrument's responsiveness. For this section of the report, we included studies that estimated an instrument's effect size or minimally important difference (MID), or that otherwise commented on the apparent ability of an instrument to assess change over time in cough frequency, cough severity, or response to a tussigenic challenge test. We also included studies that provided comparative data on two or more cough assessments before and after treatment for cough or its underlying etiology. We did not include efficacy, effectiveness, or safety trials of cough treatment strategies that did not directly or indirectly compare two or more cough assessments; such studies comprise essentially the entire published literature of interventions for cough and its underlying etiologies, as well as much of the literature on interventions with cough as a known side effect. Without a comparative analysis of two or more cough assessments as reported by individual study authors, it is difficult or impossible to determine whether a given instrument failed to detect clinical change or whether there was an absence of clinical change resulting from a given intervention.

Adults and Adolescents (≥14 Years of Age)

Thirty (83%) of the 36 studies that provided information on at least one instrument's responsiveness included adults and adolescents, and 2 studies (6%) studies included adults, adolescents, and children. Of these, eight studies reported on the LCQ, seven on the CQLQ or ACOS, and two on the CASA-Q. Three studies assessed responsiveness of tussigenic challenge tests. The remaining studies in adult and adolescent populations included information pertaining to responsiveness for a variety of different instruments, including generic health-related quality-of-life instruments, recording devices, unnamed questionnaires, and VAS and diary instruments. Below, we summarize the literature for the studies that reported responsiveness data on the LCQ, CQLQ, ACOS, and CASA-Q.

There is compelling evidence in support of the LCQ's potential ability to detect clinical change over time. Berkhof et al.106 demonstrated that the mean change in total LCQ score after 12 weeks of treatment was 4.3 (SD 2.5) among the 11 (of 49) patients who reported improvement in cough over the course of the 12 weeks. Murray et al.81 reported a change in total LCQ score from 11.3 (95% CI, 9.3 to 13.8) to 17.8 (95% CI, 14.2 to 18.8) associated with a course of antibiotic treatment for exacerbations of bronchiectasis believed to be due to a pulmonary infection. The LCQ demonstrated significant correlation with the St. George Respiratory Questionnaire over the same time period with that same group of bronchiectasis patients. Similar findings were reported by Morice et al.,80 with total LCQ scores increasing from 12.3 (SD 2.5) to 15.5 (SD 2.7) among patients whose cough was treated with morphine. Significant score increases were also noted for all three LCD domains among patients treated with morphine. Patients treated with morphine in this study also demonstrated significant reductions in cough scores as assessed by a daily cough diary, but no significant change over time was noted among patients in the placebo group. Four studies38,66,76,92 in adults and adolescents, and one study in adults, adolescents, and children69 estimated responsiveness indices for the LCQ and its domains; the values of estimates varied across the studies, in part because of different methods and reference tests.

There is also compelling evidence that suggests that the related CQLQ and ACOS instruments are responsive to clinical change. Irwin et al.67 reported strong correlation between VAS and ACOS scores over time and demonstrated that both ACOS and VAS scores changed soon after patients underwent antireflux surgery. French et al.8 demonstrated that with successful treatment of chronic cough, the average number of ACOS complaints decreased from 8.6 to 1.9 (p<0.001), and Novitsky et al.88 demonstrated that both ACOS and Symptom Inventory Profile scores correlated with improved outcomes and with each other. Field et al.56 demonstrated that mean CQLQ scores differed between patients whose cough scores (as assessed by a VAS) had disappeared, improved, or did not improve. Fletcher et al.58 demonstrated the CQLQ's responsiveness relative to a global rating of change instrument and the Punum Ladder, an instrument that allows patients to rate change in both their overall health-related quality of life associated with their cough and on six domains of quality of life associated with the six subscales of the CQLQ. French et al.60 also demonstrated that total and subscale CQLQ scores were significantly lower post-treatment when patients were no longer reporting cough as a complaint. Of the seven studies we identified that reported responsiveness-related information for the CQLQ, only one had ambiguous findings; in a 12-week RCT of proton pump inhibitors versus placebo in patients with chronic cough,94 CQLQ scores demonstrated similar changes in both treatment groups. It is possible, however, that the treatment with proton pump inhibitors was not effective, as opposed to the CQLQ not being a responsive instrument.

Two studies reported on responsiveness of the CASA-Q. Crawford et al. developed this questionnaire and validated it among patients with COPD or chronic bronchitis in Germany, France, and the U.S., using German, French, and English language versions of the questionnaire.46 This study demonstrated the responsiveness of the CASA-Q among patients who reported both worsening or improvement of cough symptoms over an unspecified period of time. Monz et al.79 estimated the mean change in CASA-Q scores over a 6-week period relative to self-reported clinical change as assessed by the Patient Global Impression of Change (PGI-C) instrument. If one considers self-reported clinical change of “a little better” (as assessed by the PGI-C) to indicate a minimally important difference (MID), the estimated MID for the CASA-Q, is 19.8 (SD 21.28). Other MID estimates for the CASA-Q among this patient population can be derived by the data reported in the paper by Monz et al.

Children (p<14 Years of Age)

Four studies (11%) reported on responsiveness of standardized cough questionnaires among children aged 14 years or younger, and two studies included adults, adolescents, and children. Three of these studies were conducted by Newcombe and colleagues84-86 on the PC-QOL. These studies included children with chronic cough and their parents. They provide compelling evidence in support of the PC-QOL's responsiveness. All subscales from the psychometric analysis showed significant improvement in parent-reported quality of life following the treatment for cough (all p<0.001), and all improvements in PC-QOL scores derived by clinical impact were significant at p<0.001. Those parents whose children had not ceased coughing reported significantly greater frequency of concern and worries on the CASA-Q than those whose children had ceased coughing. The trend for change scores were found to increase with higher VCD change ratings for the overall CASA-Q scores, as well as for its three subscales. These investigators estimated effect sizes based on clinical change scores as assessed by a VCD.

A fourth study by Fuller and colleagues61 compared parent's perception of their child's cough and sleep disturbance with cough counts through video recording. At the end of the study, 82 percent of parents correctly perceived whether their child's cough was better or worse. Most parents could not, however, comment on whether their child's sleep was disturbed.

A single study reported on responsiveness for the PCQ in children and adolescents (up to 18 years, mean age 6.8 years ± 5.1). Hartnick et al.63 developed this 5-item questionnaire and administered it to the parents of 120 children who presented to a pediatric pulmonology clinic with a chief complaint of cough. The PCQ was administered three times: prior to the first visit; within 2 weeks of the first administration but before any treatment had been instituted so that test-retest reliability could be assessed; and 3 weeks after the second administration to determine if it would accurately reflect the parent's perception of how the child's cough had changed following treatment. The authors found that mean PCQ at posttreatment was significantly different from the mean at pretreatment in cases in which parents reported that their child's cough had either improved (p<0.001) or worsened (p=0.003), whereas mean scores at these two time points were unchanged in cases in which parents reported that their child's cough was unchanged after treatment (p=0.19)

Key Question 2. Nonspecific Therapies for Chronic Cough

KQ 2: In adults and adolescents (≥14 years of age) and children (<14 years of age), what are the comparative safety and effectiveness of nonspecific (or symptomatic) therapies to treat patients with chronic cough?

  1. In patients with unexplained chronic cough
  2. In patients with refractory cough with a known underlying etiology

Key Points

Key points from the Results chapter are:

  • A wide variety of pharmaceutical agents have been used to treat the symptom of chronic cough, including opioid, anesthetic, and nonopioid/nonanesthetic antitussives; expectorant and mucolytic protussives; antihistamines; antibiotics; inhaled corticosteroids; and inhaled anticholinergics.
  • Patients with unexplained or refractory chronic cough are not well-defined as a population in the evidence base, restricting the applicability of many studies.
  • Of the agents reviewed, the opioid and certain nonopioid/nonanesthetic antitussives most frequently demonstrated efficacy for managing the symptom of chronic cough in adults.
  • There were several important quality limitations in the literature, including (1) too few good-quality studies focusing on chronic cough; (2) relatively short durations of followup 3) a diversity of outcomes measured across studies, which limited between-study comparisons; and 4) when similar outcomes were assessed across studies, the instruments used were diverse and inconsistent, making comparison and interpretation difficult.
  • Data on nonpharmacological therapies for chronic cough were sparse.
  • Studies evaluating management of unidentified or refractory chronic cough in children are extremely limited.
  • All preparations appeared to be well-tolerated, but side effects and adverse events were uncommonly reported; underreporting side effects and adverse events could limit the assessment of effectiveness of these drugs.

Description of Included Studies

Sixty-seven (67) comparisons from 48 studies evaluated therapies in patients with chronic cough and met our inclusion criteria. The 48 studies were described in 42 publications.34,80,93,96,103,108-144 Thirty-three of the 48 studies were parallel-group RCTs,34,80,93,108,109,112,115,117-120,122-125,128,130,132-135,137-144 and 12 were randomized crossover studies.96,110,111,113,114,116,121,126,127,129,131,136 The range of years of publication was 1953 to 2012; 32 (76%) of the articles were published before 2000. Only three studies130,138,144 were performed in children. The majority of studies took place in Europe, with 9 in the UK80,96,113,114,118,122,126,129,143 and 17 in other countries in Europe34,108,109,115,116,119,124,125,131,132,139,142 (total of 54%); 8 studies took place in the United States,103,110-112,123,133-135 7 in Asia,117,127,128,136,137,141 5 in Australia/New Zealand,121,130,138,140,1441 in Canada,120 and 1 in South America.93 A total of 2923 participants were included across trials; sample sizes were relatively small, ranging from 8 to 214 participants. Duration of followup was relatively short in most studies, ranging from 1 hour to 115 days. Thirty-three studies (69%) had a followup duration of 2 weeks or less. Other study characteristics are summarized in Appendix F (Table F-3). The majority of studies were rated fair in quality (n=29, or 60%); 11 studies were good in quality, and 8 were poor in quality. Fair- and poor-quality studies had the following limitations: limited description of study entry criteria, randomization, and patient population; incomplete followup; less valid statistical analyses (not intention-to-treat, post hoc subgroup analyses); and/or inadequate reporting of methods and findings.

A variety of interventions were studied; these can be broadly categorized into antitussives, protussives, and nonantitussive/nonprotussive interventions. Antitussives were further categorized as opiates, anesthetics, nonpharmacological, or “other” types. Protussives included expectorants, mucolytics, and nonpharmacological therapies. Nonantitussive/nonprotussive pharmacotherapies included antihistamines, antibiotics, anticholinergics and bronchodilators. Figure 3 represents the various classes of interventions and the comparisons among these represented in the included studies. The 48 studies represented 67 different comparisons within or between treatment classes and included studies of 59 individual agents. There were 39 comparisons (58%) with placebo. The most common class comparisons were between other antitussives and placebo (12 comparisons, 18%), followed by comparisons between antitussive opiates and placebo (11 comparisons, 16%) and comparisons between antitussive opiates and other antitussives (10 comparisons, 15%). Fourteen different class comparisons were evaluated by only one or two studies. Only two studies evaluated nonpharmacological interventions.

This figure shows the number of direct comparisons between treatments evaluated in this report. All of these treatments pertained to KQ2. The most common comparisons were between antitussive (other) and placebo (12 comparisons [18% of the total number of comparisons evaluated]), followed by antitussive (opiates) versus placebo (11 comparisons [16%]), antitussive (other) versus antitussive (opiates) (10 comparison [15%]), antitussive (other) versus antitussive (other) (7 comparisons [10%]), protussive (mucolytic) versus placebo (4 comparisons [6 percent]), and antihistamine versus placebo (3 comparisons, [4 percent]). Two comparisons (3% each) were made between protussive (expectorants) versus placebo, protussive (expectorants) versus antitussive (other), antitussive (opiates) versus antitussive (opiates), antitussive (opiates) versus antitussive (anesthetics), antibiotic versus placebo, and coritcosteroid versus placebo. A single direct comparison (1% each) was made between antitussive (anesthetics) versus placebo, antitussive (anesthetics) versus antitussive (anesthetics), antitussive (nonpharmacological) versus placebo, bronchodilator versus bronchodilator, antihistamine versus antihistamine, protussive (nonpharmacological) versus protussive (nonpharmacological), protussive (mucolytic) versus protussive (mucolytic), and anticholinergic versus placebo.

Figure 3

Overview of intervention class comparisons.

Within the included studies, a variety of causes for chronic cough were represented, including bronchitis, chronic obstructive pulmonary disease (COPD), asthma, upper airway cough syndrome (UACS), fibrosis, neoplasm, tuberculosis, cystic fibrosis, gastroesophageal reflux disease (GERD), and unexplained cough, among others (Appendix F, Table F-3).

Table 8 details the specific agents used within the different class comparisons. It also lists the categories of outcomes assessed. The most frequent outcomes studied were cough severity/QOL (57 comparisons, 85%) and cough frequency (37 comparisons, 55%). However, even within these outcome categories, the instruments used to measure similar outcomes varied widely across studies. Other outcomes included functional status, tussigenic challenge, global assessment, and adverse effects, among others.

Table 8. Interventions and outcomes assessed.

Table 8

Interventions and outcomes assessed.

Detailed Synthesis

Table 9 summarizes the patient-centered outcomes data collected for each study.

Table 9. Patient-centered outcomes data.

Table 9

Patient-centered outcomes data.

Studies Involving Opiate Antitussives

Among the studies reviewed, we found 29 comparisons within 17 studies involving opiate antitussives: 11 comparisons were with placebo,80,96,103,110-112,116,125,132,133,135 2 comparisons were of different doses of the same opiates,132,133 2 comparisons were with anesthetic antitussives,117,135 and 10 comparisons were with nonopioid/nonanesthetic antitussives.34,108,110-112,116,119,124,125 In the 11 comparisons of opiates with placebo,80,96,103,110-112,116,125,132,133,135 opiates were more effective for improving cough frequency, cough severity, and/or quality of life (LCQ) in 8 of the studies.80,103,110-112,116,125,133 The effective regimens in these studies were codeine 7.5–60 mg or morphine 5 mg. In one study, the opioid morphine resulted in significant rates of constipation and drowsiness but was not discontinued due to tolerability issues.80 No one opioid was found to be superior to another in within-class comparisons, although codeine had a dose-response improvement in cough severity and frequency in a study comparing the following doses: 7.5 mg, 15 mg, 30 mg and 60 mg.133 A study of another opiate, viminol, found a higher dose (140 mg) to be effective at reducing cough severity, but a lower dose (70 mg) was no different from placebo.132

When compared with anesthetic antitussives, opiates were not more effective for cough in the two studies making this comparison.117,135 Opiates (only codeine derivatives in these studies) were compared with a variety of nonopioid/nonanesthetic antitussives. Codeine had comparable efficacy for reducing cough frequency, but was less effective than dextromethorphan for improving cough severity in one study.125 In another study, the two agents were comparable for cough severity.112 Two studies showed codeine 15–17 mg 3–4 times a day was more effective at reducing cough severity than low-dose dextromethorphan (4-6 mg 3–4 times a day).110,111 Two studies comparing codeine with glaucine reported conflicting results; one noted significantly better improvement in cough severity and frequency with glaucine,119 while the other noted no significant differences in cough frequency over 8 hours of followup; codeine did result in lower frequency of cough over the final 2 hours of the 8-hour duration.116 Codeine or dihydrocodeine were similar in effectiveness for cough to moguisteine,34 levodropropizine,124 and levocloperastine108 in studies making these comparisons.

In terms of tolerability, 2 of 39 patients taking codeine 30 mg in one study discontinued the study medication due to side effects: dry mouth and asthenia in one patient, nausea in the other patient.34 In another study, the percentage of patients experiencing somnolence while taking dihydrocodeine was significantly higher (22%) than in in the group receiving levodropropizine (8%).124 In two studies, side effects of nausea, constipation, and/or drowsiness were more frequent with codeine than with dextromethorphan.110,111

Studies Involving Anesthetic Antitussives

Anesthetic antitussives were examined in three studies resulting in four comparisons: one comparison was with placebo,135 two were with opioids,117,135 and one study compared two anesthetic antitussives.134 Benzonatate was not superior to placebo in the one study making this comparison.135 Clophedianol and benzonatate were not more effective for cough than opiates in the two studies making these comparisons.117,135 The two anesthetic antitussives, benzonatate and Becantyl® (sodium 2.6 ditertiarybutylnaphtalene monosulphonate; also named Becantex® or L.1633), had comparable effects on cough severity in one study.134

Studies Involving Other Antitussives

We found 31 comparisons in 21 studies involving other (i.e., nonopioid/nonanesthetic) antitussives: 12 were comparisons with placebo,109-112,116,125,127,129,131,137,142 7 were comparisons between 2 nonopioid/nonanesthetic antitussives,108,112,115,131 1 was a comparison of a combined antitussive/expectorant with the expectorant alone,142 and 10 were comparisons with opioids34,108,110-112,116,119,124,125 (see the section on “Studies Involving Opiate Antitussives” for a summary of these studies). Six studies compared dextromethorphan with placebo,110-112,125,129,131 with 5 of these supporting the efficacy of dextromethorphan for treatment of cough.110-112,125,129 In one study,125 dextromethorphan 20 mg twice a day was more effective than placebo for reducing both cough frequency and severity. In another study,129 dextromethorphan 30 mg three times a day was more effective than placebo at reducing cough in response to citric acid tussigenic challenge but not for cough severity, sleep disturbance, or cough-specific quality of life (LCQ). In three studies, dextromethorphan 4–20 mg 3–4 times a day were better than placebo for improvement of cough severity.110-112 In the one negative study,131 dextromethorphan 30 mg once was no better than placebo for impacting cough frequency or severity. A study examining glaucine 30 mg compared with placebo noted improvements in a 6-hour subset of 8-hour nocturnal cough frequency, but not in the full 8-hour duration of followup.131 Another study of glaucine 30 mg noted lower cough frequency than placebo after 4 hours.116 A Chinese herbal medicine consisting of extracts from six crude herbs called bakumondoto (3 g of powder before each meal) reduced cough severity and frequency compared with no treatment control over 8 weeks by participant self-report in diaries and on a visual analog scale.127 Moguistene 200 mg 3 times a day was more effective than placebo for reducing cough frequency over 4 days.109 Pipazethate 20 mg did not reduce cough frequency compared with placebo.137

In seven studies, none of the “other” antitussives was found to be superior to another in comparisons made among them,108,115,131 nor were different doses of the same agent.112

One study compared the combination antitussive/expectorant Duopect® (narcotine/glycerol) 17 mg/120 mg 3 times a day, narcotine 17 mg, glycerol 120 mg 3 times a day, and placebo. More participants taking Duopect or narcotine than glycerol alone or placebo reported moderate to marked improvement in cough severity (n=28 vs. 24 vs. 13 vs. 3, respectively, of 32 patients in each arm, p<0.01 for Duopect and narcotine vs. glycerol and placebo).142 Expectoration was “easier and freer” in a higher percentage of patients taking Duopect or glycerol than narcotine or placebo. In a multi-arm study, dextromethorphan 10 mg and 20 mg were similarly better than placebo for improvement of cough severity.112

Studies Involving Protussives

There were 10 comparisons involving protussives: two were comparisons of an expectorant with placebo,128,142 two were comparisons of an expectorant with an ‘other’ antitussive,126,142 four were comparisons of a mucolytic with placebo,118,120,122,128 and one study compared a mucolytic with another mucolytic.114 Compared with placebo, the expectorant bromhexine did not reduce cough frequency in one study.128 In another study, guaifenesin reduced cough intensity (on 1 of 4 days that were analyzed over 2 weeks), increase sputum volume (on 1 of the 4 days), and improved subjective “ease of expectoration” in a subgroup of high-volume sputum patients.128 Another study compared the combination antitussive/expectorant Duopect (narcotine/glycerol) 17 mg/120 mg 3 times a day, narcotine 17 mg, glycerol 120 mg 3 times a day, and placebo. More participants taking Duopect or narcotine than glycerol alone or placebo reported moderate to marked improvement in cough severity (n=28 vs. 24 vs. 13 vs. 3, respectively, of 32 patients in each arm, p<0.01 for Duopect and narcotine vs. glycerol and placebo).142 Expectoration was “easier and freer” in a higher percentage of patients taking Duopect or glycerol than narcotine or placebo. The following regimens did not improve cough frequency, severity, and/or quality of life compared with placebo: N-acetylcysteine 200 mg 3 times a day,122 N-acetylcysteine 4 mg inhaled via metered-dose inhaler 4 times a day,118 bromhexine 16 mg 4 times a day.128 or ambroxol 60 mg twice a day.120 In a comparison of inhaled aerosols, 2-mercapto-ethane sulfonate was not more effective than hypertonic saline at improving tracheobronchial clearance.114

Studies Involving Nonantitussive and Nonprotussive Pharmacotherapies

We identified four studies examining the effect of antihistamine medication on cough123,130,136,138 Two of these studies were in children130,138 and are discussed separately below. In the adult studies, one compared diphenhydramine 50 mg with diphenhydramine 25 mg to placebo, all preparations scheduled 4 times a day.123 The two doses of diphenhydramine did not differ in efficacy for cough frequency, but both doses were superior to placebo. Higher dose diphenhydramine resulted in a greater frequency of drowsiness than the lower dose, which was comparable with placebo. In a placebo-controlled study,136 loratadine 10 mg reduced the number of coughs following tussigenic challenge with ultrasonically nebulised distilled water in patients with nasal disease or unexplained chronic cough but not in normal patients.

One study examined the effect of the antibiotic erythromycin at a dose of 250 mg once a day and found no difference in cough severity, cough frequency, cough-specific quality of life or response to tussigenic challenge compared with placebo.143 A study compared ipratroprium bromide 20 mcg inhaler, 4 puffs 4 times a day, with placebo and found improvements in cough severity and dyspnea associated with cough.121 A study comparing two bronchodilators (diprophylline 200 mg 3 times a day vs. methoxyphenamine 2 capsules 3 times a day) did not find differences in their effects on cough frequency or cough-specific quality of life.141 Two studies evaluated inhaled corticosteroids in patients with chronic cough of unknown etiology and found that cough severity was reduced113 or cough more frequently resolved93 compared with placebo.

Studies Involving Nonpharmacological Therapies

We identified only two studies that evaluated the comparative safety and effectiveness of nonpharmacological interventions for chronic cough.139,140 Van Hengstum et al.139 compared 20 minutes of positive expiratory pressure (PEP) physiotherapy with 30 minutes of a forced expiratory technique (FET) and no treatment using a randomized crossover trial involving eight adult patients (age range, 48–73 years) with chronic bronchitis. FET was found to be more effective than either PEP or no treatment in enhancing the primary outcome of tracheobronchial clearance, but there was no evidence that either treatment was effective in improving cough frequency or severity. This study was rated as fair quality because of the small sample size and nonblinded study design. Applicability is limited due to incomplete reporting of the interventions and the use of short-term, surrogate outcomes.

The second study was a randomized, placebo-controlled trial that compared speech pathology management with placebo among 87 adult patients with refractory chronic cough of at least 2 months in duration.140 Patients in both study arms participated in four individual 30-minute intervention sessions with a speech pathologist with experience in treating voice disorders. The active intervention included targeted education and training in strategies to reduce cough and laryngeal irritation. The placebo intervention consisted of healthy lifestyle education, stress management, exercise, and diet. Patients in the intervention arm demonstrated greater reduction in cough (p=0.003) and limitation of symptoms on everyday activity (p=0.011) symptom scores relative to those in the placebo arm. The active treatment was also associated with greater reduction in breathing, voice, and upper airways symptom scores relative to the placebo intervention. This study was rated as fair quality because of the single-blind study design and the lack of a validated outcome measure. Applicability is limited by an intervention that requires a level of training or proficiency that is not widely available.

Studies Involving Children

Three studies addressed the treatment of chronic cough in children less than 14 years of age. Two of these evaluated ketotifen, an H1-antihistimine and mast-cell stabilizer.130,138 In the United States, ketotifen is currently not available in oral form but is available as an eye drop for allergic conjunctivitis. The oral form is, however, available internationally, including from Canada. Both studies were RCTs of ketotifen versus placebo for children with chronic cough and/or wheeze. One evaluated 113 children between 6 and 36 months of age over 16 weeks,138 and the other 214 children between 2 and 6 years of age over 12 weeks.130 In the study of younger children, ketotifen was not more efficacious than placebo. However, the study of older children reported that the number of exacerbations of cough and wheeze lasting 3 or more days was reduced in the group treated with ketotifen compared with placebo. In addition, there was a decrease in the proportion of children taking beta-agonists and methylxanthines. The study of younger children was rated as good quality. The study of older children was rated fair because there was no allocation concealment, the primary outcome measure was unclear with multiple comparisons, and the study was industry funded. These two studies have low applicability to the management of children with chronic cough. In both studies, all subjects likely had asthma as their source of chronic cough. These studies were published over 20 years ago (in 1989 and 1992). The management of asthma has significantly changed since these studies were conducted, with greater emphasis on the role of controller medicines (e.g., inhaled corticosteroids, leukotriene inhibitors) to reduce the chronic symptoms associated with poorly controlled asthma. It is unclear whether findings regarding ketotifen are generalizable to the other available medications in its class.

The third study was a randomized, placebo-controlled trial of an antibiotic, amoxicillin clavulanate, in children with more than 3 weeks of wet cough.144 Children were randomized to 14 days of antibiotic or placebo, and outcomes were measured with a cough diary using the verbal category descriptive score. The primary outcome was cough resolution, based on at least 75 percent reduction in cough score on average during the 2 days following treatment or 3 days within a the trial period. Fifty children were enrolled. The mean age in the treatment group was 1.75 years (range, 0.9 to 4.6 years) and 2.8 years (range, 0.95 to 5.25 years) in the control group. Cough resolution was 48 percent in the treatment group and 16 percent in the placebo group (p=0.0016), with a number needed to treat of 4. This study was rated as good quality, and although it had a small sample size and the description of diagnostic evaluation of cough was minimal, it otherwise had good applicability.

Quantitative Synthesis

The heterogeneity of the included studies in terms of the interventions and comparators (Figure 3), combined with the lack of three or more studies reporting the same outcome where there were multiple comparisons (Table 8), precluded us from performing meta-analyses on almost all outcomes.

We were, however, able to evaluate the relative effects on cough severity for four classes of treatments for chronic cough: antitussive opiates, antitussive dextromethorphan, antitussive moguisteine, and protussive mucolytics. Thirteen studies reported results for cough severity, but two of these116,131 did not provide sufficient information to estimate effect sizes. Of the remaining 11 studies (n=396 patients), 4 provided information on mucolytics, 3 provided information on dextromethorphan, 2 provided information on moguisteine, and 4 provided information on opiates. Most of the 11 studies compared the treatment with placebo,34,80,115,118,120,122,125,128,129,132,135 but one compared opiates with dextromethorphan and placebo.125 Methods used to measure cough severity differed widely amongst the studies, from studies looking at the proportion of patients receiving good or excellent cough relief after treatment,134 to those evaluating a mean cough severity score using various Likert scores or VAS instruments,34,80,115,120,122,129,132 to those measuring the median or mean change in intensity of cough.118,125,128 Because each study used a different measure of severity, we converted all results to effect sizes (standardized mean differences). Relative to placebo, the effect of dextromethorphan on cough severity was 0.54 (95% confidence interval [CI], 0.27 to 0.80; p=0.0008), the effect of opiates was 0.63 (95% CI, 0.40 to 0.86; p<0.0001), the effect of moguisteine was 0.62 (95% CI, 0.04 to 1.16, p=0.0366), and the effect of mucolytics was 0.14 (95% CI -0.20 to 0.49; p=0.384; Figure 4). The studies showed heterogeneity (p=0.0023). The effects of dextromethorphan, moguisteine, and opiates compared with placebo on cough severity support a benefit of these therapies, but the evidence is insufficient to determine relative benefit among these therapies.

This figure is a forest plot showing the relative effects on cough severity of antitussive opiates, the antitussive dextromethorphan, and protussive mucolytics. Effect sizes relative to placebo (with 95% CIs) are: for dextromethorphan, 0.54 (95% CI, 0.27 to 0.80; p=0.0008); for opiates, 0.63 (95% CI, 0.40 to 0.86; p<0.0001); for moguisteine, 0.62 (95% CI, 0.04 to 1.16, p=0.0366); and for mucolytics, 0.14 (95% CI -0.20 to 0.49; p=0.384). The studies showed heterogeneity (p=0.0023). The effects of dextromethorphan, moguisteine, and opiates compared with placebo on cough severity support a benefit of these therapies, but the evidence is insufficient to determine relative benefit among these therapies.

Figure 4

Meta-analysis of data on cough severity.

We performed a similar meta-analysis for cough frequency, including 7 studies (n=700 patients).34,109,115,116,125,133,135 Relative to placebo, the effect of dextromethorphan on cough frequency was 0.40 (95% CI, 0.18 to 0.85; p=0.0248), the effect of codeine was 0.57 (95% CI, 0.36 to 0.91; p=0.0260), and the effect of moguisteine was 0.60 (95% CI, 0.31 to 1.17, p=0.1117; Figure 5). The studies showed significant heterogeneity (p=0.0231). The effects of dextromethorphan and codeine compared with placebo on cough frequency support a benefit of these therapies, although the estimates are too imprecise to determine if one is superior to another. The effect of moguisteine was too imprecise to draw conclusions about is efficacy.

This figure is a forest plot showing the relative effects on cough frequency of antitussive opiates, the antitussive dextromethorphan, and protussive mucolytics. Effect sizes relative to placebo (with 95% CIs) are: for dextromethorphan, 0.40 (95% CI, 0.18 to 0.85; p=0.0248); for codeine, 0.57 (95% CI, 0.36 to 0.91; p=0.0260); and for moguisteine, 0.60 (95% CI, 0.31 to 1.17, p=0.1117). The studies showed significant heterogeneity (p=0.0231). The effects of dextromethorphan and codeine compared with placebo on cough frequency support a benefit of these therapies.

Figure 5

Meta-analysis of data on cough frequency.

Views

  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (2.4M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...