This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lee, P. C. L. Articles by Eccles, R. Search for Related Content PubMed PubMed Citation Articles by Lee, P. C. L. Articles by Eccles, R. Related Collections Pulmonary Therapeutic Interventions

The Antitussive Effect of Placebo Treatment on Cough Associated With Acute Upper Respiratory Infection

From the Common Cold Centre (P.C.L.L., M.S.M.J., R.E.), Cardiff School of Biosciences, Cardiff University, Cardiff, UK; and Procter and Gamble Technical Centres Limited (J.D.H., W.H.L.W., K.S.), Surrey, UK.

Address correspondence and reprint requests to Professor Ron Eccles, Common Cold Centre, Cardiff School of Biosciences, Cardiff University, Cardiff CF1 3US, UK. E-mail: eccles{at}cardiff.ac.uk

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION NOTES REFERENCES

 
Objective: The objective of this study was to determine the effects of a placebo treatment on cough in patients with cough associated with acute upper respiratory tract infection (URTI).

Methods: Patients with dry or slightly productive cough associated with a history of URTI were recruited. Cough frequency (CF) over 15 minutes was recorded by means of a microphone connected to a pen recorder. Cough suppression time (CST) was recorded when patients were instructed by means of a red light to try not to cough. Patients received either a single dose of vitamin E (placebo treatment) or no treatment. CF and CST were recorded before and 15 minutes after treatment.

Results: Twenty-seven patients were randomized to placebo treatment and 27 to the no-treatment group (mean age 22.6 years). The median difference between post- and pretreatment CF was –3 in the no-treatment group and –18 in the placebo group (p = .0003). There was a significant increase in CST in the placebo group compared with no treatment (p = .027).

Conclusions: The results demonstrate that placebo treatment has significant antitussive activity. This placebo effect may be related to generation of central neurotransmitters such as endogenous opioids.

Key Words: cough • common cold • placebo effect

Abbreviations: CF = cough frequency; CST = cough suppression time.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION NOTES REFERENCES

 
To determine the antitussive efficacy of the specific pharmacologic component of a cough medicine, clinical trials are designed to compare the efficacy of the cough medicine with a matched placebo. Studies on the effects of antitussive medicines on cough associated with acute upper respiratory tract infection (URTI) have often failed to demonstrate any superiority of the cough medicine above placebo because of the large antitussive effect of the placebo treatment (1–3). In 1990, Fuller stated that "the reason for any efficacy of proprietary cough medicines containing opiates probably lies in the sugar solution in which they are prepared as only limited pharmacological activity has been detected at these doses" (4). However, proprietary cough medicines do have considerable antitussive activity. This is probably related to a combination of the placebo effect of the medicine and the pharmacologic activity of the specific pharmacologic component. In clinical trials on cough associated with URTI, it is difficult to measure the efficacy of the placebo component of treatment, because there may be some tendency for spontaneous resolution of cough during the course of the study. A marked reduction in cough frequency associated with placebo treatment has been previously explained to be the result of resting the patients (1).

The objective of the present study was to determine the effects of a placebo treatment on cough in patients with cough associated with acute URTI. Two measures of cough are used in the present study; cough frequency (CF) and cough suppression time (CST, the time in minutes that a patient could suppress their cough), because it was proposed that they may measure different types of cough. A component of cough associated with URTI has been shown to be suppressible by voluntary control (5). It was therefore proposed that cough associated with URTI may consist of two separate components, suppressible and nonsuppressible cough, and that CST would provide a measure of nonsuppressible cough, whereas CF would provide a measure of combined suppressible and nonsuppressible cough. If nonsuppressible cough is not under any significant voluntary control, then one might expect CST to be unaffected by treatment with placebo.

To determine the magnitude of any placebo treatment on cough, it is necessary to compare placebo treatment with "no treatment" because this will control for any regression to mean or natural decrease in cough severity. Therefore, a "no-treatment" group was used for comparison with a placebo treatment group (vitamin E) in the present study.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION NOTES REFERENCES

 
Patients
Patients aged between 18 to 65 years, with dry or slightly productive cough associated with a history of acute URTI of more than 1 day but less than 14 days duration, were recruited from the staff and students of Cardiff University and from the general population of the city of Cardiff. Patients were excluded from the study if they had a history or current condition of asthma or lower respiratory tract infections, had a clinically significant disease or history of disease that was deemed to be likely to affect their participation in the study, were lactating or pregnant, had taken a product containing menthol in the previous 6 hours or any other medication in the past 24 hours that was deemed to be contraindicated for the study (eg, antitussives, antihistamines), or had a history of allergy or any adverse reaction to vitamin E. Smokers were not excluded from the study as the University population had few smokers, and these were not heavy smokers. All patients gave written informed consent for the study, which was approved by the Bro Taf Area Heath Authority Ethics Committee.

Protocol
The study was designed to investigate the effects of a placebo treatment on cough, and the wording of the information provided to the patients was critical, because this needed to be accurate but not misleading. Because of the negative connotation associated with the term "placebo," this term was not used in the patient information. Patients were given the following information as well as information about the sponsor, adverse effects, and indemnity:

"The treatment to be tested contains vitamin E 300iu."

"The study is designed to investigate the effect of vitamin E on cough associated with the common cold."

"Depending on the frequency of your cough, you will be allocated to receive either no treatment or vitamin E. If you are in the treatment group, you will be given a capsule to take with a cup of water and you will be asked to swallow the capsule with all of the water provided and not to chew it. If you are in the no-treatment group, you will be instructed to drink all of the water in the cup provided without taking any medication."

"This clinical trial is concerned with finding out more about the control of cough and you may not receive any clinical benefit by participating in the study."

Patients were seated comfortably in a quiet room in front of three colored lights with the following instructions placed near each light: red light (please do not cough), green light (just relax and cough if you wish), or white light (please cough once only). The lights were controlled by an operator, who sat in a screened area out of sight of the patient. The protocol was explained to the patient, and they were asked to follow the instructions as indicated by the lights being switched on and off. The protocol consisted of a pretreatment cough-recording session (31 minutes), a treatment session (15 minutes), and a posttreatment recording session (31 minutes). Each cough-recording session comprised 15 minutes of free cough (green light), two voluntary coughs at 0.5-minute intervals (white light), and 15 minutes of cough suppression (red light). The two voluntary coughs were obtained at the start of each period of cough suppression to ensure that all patients started the suppression period under similar conditions. The recording procedures are summarized in Table 1.

Medication
Placebo Treatment Group
A single dose of 300 IU (201 mg) vitamin E in an oil base, in small, yellow and oval gelatin capsules (Superdrug Plc, Croydon, UK) was chosen for this study as a placebo treatment. This medication was chosen because there is no evidence for any effect of vitamin E on cough, and because the absorption of vitamin E from the gut has a lag time of 2 to 4 hours, ensuring that there could be no physiological or pharmacologic effects of the vitamin on cough as measurements were made 15 minutes after dosing (6). However, the patient might consider it reasonable that a "vitamin" could have some effect on cough. Patients were asked to swallow the medication with 50 mL of still water.

No-Treatment Group
Patients were informed that they had been allocated to receive no treatment for their cough and were asked to drink 50 mL of still water.

Cough Recording
Cough frequency was recorded by means of a microphone connected to a pen recorder (Lafayette Datagraph 76100B ink pen recorder) using the integrator channel of an EMG amplifier (Lafayette 76409) with the time constant set at 0.02 sec (5). The microphone was mounted on a stand and placed on the floor in front of the patient. The pen recorder was placed behind a screen and observed by an investigator who also controlled the colored lights. The sensitivity of the pen recorder amplifier was adjusted for each patient over a 5-minute period before the recording sessions so that maximum pen deflections were produced when the patient was coughing. To set the sensitivity of the pen recorder, the patients were asked to make two or three voluntary coughs. The mean (in millimeters) of the three largest pen deflection for that patient during the pretreatment cough session was calculated and any cough deflection, which was greater than one third of this mean maximum deflection was counted as a cough (5). Cough frequency was expressed as the number of coughs per 15-minute recording period.

Statistical Methods
This was a pilot study and the number of patients was dependent on the incidence of URTI during the common cold season in the autumn–spring period. A conventional power calculation was considered inappropriate, because no comparable previous study was available. From previous experience with studies of cough associated with URTI, it was proposed to recruit a minimum of 40 patients (ie, 20 per treatment group) up to a maximum of 80 patients.

Cough Frequency
CF was deemed to be the number of coughs recorded over 15 minutes and was expressed as the median with interquartile range. The Wilcoxon test was used to test the hypothesis that there was no change in cough frequency between the pretreatment and posttreatment measurements within the two treatment groups. The Mann-Whitney U test was used to test the hypothesis that there was no difference between the two treatment groups for changes in cough frequency between the pretreatment and posttreatment measurements. Ninety-five percent confidence intervals are given with the p value.

Cough Suppression Time
CST is given as the mean value in minutes with the standard error of the mean. Analysis of variance (ANOVA) was considered the principal statistical model to analyze changes in CST (CST posttreatment–pretreatment). The analysis of CST data were undertaken as a form of survival analysis using ANOVA as the principle statistical model, and because the protocol only allowed a maximum CST of 15 minutes, CST equal to 15 minutes was considered as censored data. A dummy variable representing censored data pretreatment, and a dummy variable representing censored data posttreatment was included in the model. Ninety-five percent confidence intervals are given with the p value.

SPSS 9.0 and Minitab 12 and SAS statistical packages were used for statistical analysis.

Outcomes
This was a pilot study and primary and secondary outcome measures were not predefined but the measures of interest were the difference between the changes in post- and pretreatment CF for the treatment groups and similarly for CST.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION NOTES REFERENCES

 
Patients
Fifty-five patients with dry or slightly productive cough associated with a history of recent URTI were recruited for the study. Fifty-four patients (27 males and 27 females, mean age 22.6 years with a range of 18–56) were included, and one patient with asthma was excluded. Twenty-seven patients were randomized to the placebo treatment group and 27 to the "no-treatment" group. All 54 patients that were treated completed the study and the data analysis relates to the 27 patients in each treatment group. No adverse effects were reported in the study.

Cough Frequency
The change in CF (posttreatment –pretreatment) for each patient is illustrated in Figure 1. The figure shows that there was a reduction in cough frequency in 19 of 27 patients in the no-treatment group and 26 of 27 patients in the placebo treatment group. The magnitude of the reduction in cough frequency was greater in the placebo group than the no-treatment group as illustrated in Figure 2, which shows the median changes in each treatment group. The CF declined from a median of 41.0 (20.0–63.0) to 38.0 (23.0–49.0) in the no-treatment group (p = .008; 1.0–9.0) and from 37.0 (23.0–59.0) to 19.0 (10.0–28.0) in the placebo group (p <.001; 13.5–27.0). The median difference between post- and pretreatment cough frequency was –3 in the no-treatment group and –18 in the placebo group (p = .0003; –7––21). A minus sign before the cough frequency value indicates a reduction in cough frequency between pre- and posttreatment recordings.

View larger version (18K): [in this window] [in a new window]   Figure 1. Changes in cough frequency per 15 minutes (posttreatment –pretreatment) for each patient in the "no-treatment" and placebo treatment groups.

View larger version (10K): [in this window] [in a new window]   Figure 2. Median cough frequency (per 15 minutes) pretreatment and posttreatment. Open symbols represent the "no-treatment group" and filled symbols the "placebo treatment group."

Cough Suppression Time
The majority of patients (38 of 54) were able to suppress their cough for 15 minutes in the pretreatment period and similarly in the posttreatment period (39 of 54). The mean difference in CST, post- minus pretreatment was 0.03 (0.49) minutes within the no-treatment group and 1.427 (0.45) minutes within the placebo group. The mean difference in CST between treatment groups was (post –pre) 1.40 minutes (p = .027; 0.16–2.63).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION NOTES REFERENCES

 
The results demonstrate that placebo treatment with a single dose of 300 IU vitamin E medication caused a significant reduction in cough frequency (50%) and a significant increase in CST compared with a no-treatment group. The no-treatment group also showed a statistically significant decrease in CF after 15 minutes, although the magnitude of this change was only 7%, and may be the result of rest causing a small decrease in cough severity.

The large decrease in CF caused by placebo treatment compared with the relatively small decrease in CF in the no-treatment group supports the hypothesis that placebo treatment has some antitussive activity over and above that which could be expected from simply resting the patients. Because of the short time period between treatment administration and posttreatment measurement of cough, and the 2- to 4-hour lag period required for absorption of vitamin E (6), vitamin E can be considered a placebo treatment.

The significant increase in CST caused by placebo treatment was not expected. The hypothesis to be tested in the present study was that CST was a measure of nonsuppressible cough and that this type of cough would not be susceptible to placebo treatment because it was not under voluntary control. The difference in CST between the no-treatment and placebo groups was only 1.4 minutes, and this may be a statistical anomaly as a result of the small sample size. However, if this finding can be repeated in other studies, it indicates that the antitussive activity associated with placebo treatment may not be solely explained by a voluntary effort to reduce cough. The degree of voluntary suppression of cough has been previously reported to be related to the severity of cough with an inverse relationship between the CSF and the baseline frequency of cough (5). Although both CST and CF may be used as measures of cough severity, they may measure different components of cough and may respond differently to antitussive medications. In this study, the cough measurements were made under laboratory conditions and the results may not necessarily be transferable to a home environment or "real-life" situation.

A large placebo effect has been previously demonstrated in similar studies on the effects of antitussive medications on cough associated with URTI. In a study on the effects of codeine (codeine syrup B.P. 30 mg/10 mL) on cough associated with URTI, Eccles et al. (1) reported a 44% reduction in CF (from 1.47 to 0.77 coughs per minute) associated with placebo treatment. In a study on cough associated with URTI, using a single dose of 50 mg codeine phosphate in a gelatin capsule or matched capsule as placebo treatment, Freestone et al. (2) reported a 50% reduction in CF (from 25.4 to 12.71 coughs per 10 minutes) associated with the placebo treatment. In both of these studies, the reduction in CF associated with placebo treatment was similar to that caused by codeine, and no difference could be shown between the two treatment groups. Similar large placebo effects, with a 37% to 44% reduction in CF, have been shown in other studies on cough associated with URTI (7). In the present study, no treatment (that is, no treatment except for ingestion of 50 mL of still water), was associated with a 7% reduction in CF, whereas treatment with placebo caused a 44% reduction in CF. The reduction in CF found in the present study associated with placebo treatment is similar in magnitude to that reported in other studies as detailed previously.

The magnitude of effect of the placebo treatment may be influenced by many factors such as belief about the medication, conditioning, the environment in which the medication is administered, the person administering the medication, and so on, and there is a growing literature on the efficacy of placebo treatment for a wide range of illnesses (8). Placebo responses have been reported to contribute between 65% to 80% of the response to antidepressant medications, indicating that any specific pharmacologic component of antidepressant medical treatment may contribute only a small fraction to the total benefit of antidepressant therapy (9).

The effects of placebo treatments have been most studied in relation to analgesia, and there is a current hypothesis that endogenous opioids contribute to placebo analgesia treatment (10,11). Evidence that placebo treatment activates the endogenous opioid system comes from studies on pain, in which it has been reported that the opioid antagonist naloxone blocks the placebo analgesic response (12). Placebo treatment has also been shown to cause a respiratory depressant response in patients after repeated administration of an opioid agonist, and this placebo effect on respiration has been proposed to be mediated by endogenous opioids, because it is antagonized by naloxone (13).

The hypothesis that some of the effects of placebo treatment may be mediated by endogenous opioids may explain the results of the present study, in which a placebo treatment has been shown to reduce CF and prolong CST. The evidence in the literature for the antitussive efficacy of opioids such as codeine is mainly supported by studies on anesthetized animals, and induced and chronic cough models in man (14), with little objective support for any antitussive efficacy for cough associated with URTI (1). The difficulty in demonstrating antitussive activity in clinical trials on cough associated with URTI may be partly explained by the great variability of this type of cough.

The results of the present study provide some new knowledge about the effects of placebo treatment on cough and indicate that the placebo effect is a major component of any cough treatment.

	NOTES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION NOTES REFERENCES

 
This research was supported by Procter and Gamble Technical Centres Ltd., Rusham Park, Whitehall Lane, Egham, UK.

Received for publication July 12, 2004; revision received September 9, 2004.

DOI:10.1097/01.psy.0000155667.59662.92

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION NOTES REFERENCES

 

1. Eccles R, Morris S, Jawad M. Lack of effect of codeine in the treatment of cough associated with acute upper respiratory tract infection. J Clin Pharmacol Ther 1992;17:175–80.[Medline]
2. Freestone C, Eccles R. Assessment of the antitussive efficacy of codeine in cough associated with common cold. J Pharm Pharmacol 1997;49:1045–9.[Medline]
3. Lee PCL, Jawad MSM, Eccles R. Antitussive efficacy of dextromethorphan in cough associated with acute upper respiratory tract infection. J Pharm Pharmacol 2000;52:1137–42.[CrossRef][Medline]
4. Fuller RW, Jackson DM. Physiology and treatment of cough. Thorax 1990;45:425–30.[Medline]
5. Hutchings HA, Eccles R, Smith AP, Jawad M. Voluntary cough suppression as an indication of symptom severity in upper respiratory tract infections. Eur Respir J 1993;6:1449–54.[Abstract]
6. Ferslew K, Acuff R, Daigneault E, Woolley T, Stanton P. Pharmacokinetics and bioavailability of the RRR and all racemic stereoisomers of alpha-tocopherol in humans after single oral administration. J Clin Pharmacol 1993;33:84–8.[Abstract]
7. Parvez L, Vaidya M, Sakhardande A, Subburaj S, Rajagopalan TG. Evaluation of antitussive agents in man. Pulmonary Pharmacology 1996;9:299–308.[CrossRef][Medline]
8. Harrington A, ed. The Placebo Effect. An Interdisciplinary Approach. Cambridge, MA: Harvard University Press; 1999.
9. Kirsch I. Are drug and placebo effects in depression additive? Biol Psychiatry 2000;47:733–5.[Medline]
10. Fields HL, Price DD. Towards a neurobiology of placebo analgesia. In: Harrington A, ed. The Placebo Effect. An Interdisciplinary Approach. Cambridge, MA: Harvard University Press; 1999:93–116.
11. ter Riet G, de Craen AJM, de Boer A, Kessels AGH. Is placebo analgesia mediated by endogenous opioids? A systematic review. Pain 1998;76:273–5.[CrossRef][Medline]
12. Benedetti FAM. The neurobiology of placebo analgesia: from endogenous opioids to cholecystokinin. Prog Neurobiol 1997;52:109.[Medline]
13. Benedetti F, Amanzio M, Baldi S, Casadio C, Maggi G. Inducing placebo respiratory depressant responses in humans via opioid receptors. Eur J Neurosci 1999;11:625–31.[CrossRef][Medline]
14. Eddy NB, Friebel H, Hahn KJ, Halbach H. Codeine and its alternates for pain and cough relief. Bull World Health Organ 1969;40:425–54.[Medline]

This article has been cited by other articles:

				B. S. Oken Placebo effects: clinical aspects and neurobiology Brain, June 21, 2008; (2008) awn116v1. [Abstract] [Full Text] [PDF]

				S. B. Mazzone, L. McLennan, A. E. McGovern, G. F. Egan, and M. J. Farrell Representation of Capsaicin-evoked Urge-to-Cough in the Human Brain Using Functional Magnetic Resonance Imaging Am. J. Respir. Crit. Care Med., August 15, 2007; 176(4): 327 - 332. [Abstract] [Full Text] [PDF]

				F. G. Miller, D. Wendler, and R. Eccles PLACEBO RESEARCH AND THE SPIRIT OF INFORMED CONSENT * RESPONSE Psychosom Med, July 1, 2005; 67(4): 678 - 678. [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lee, P. C. L. Articles by Eccles, R. Search for Related Content PubMed PubMed Citation Articles by Lee, P. C. L. Articles by Eccles, R. Related Collections Pulmonary Therapeutic Interventions