Effect of nasal balloon autoinflation in children with otitis media with effusion in primary care: an open randomized controlled trial
======================================================================================================================================

* Ian Williamson
* Jane Vennik
* Anthony Harnden
* Merryn Voysey
* Rafael Perera
* Sadie Kelly
* Guiqing Yao
* James Raftery
* David Mant
* Paul Little

## Abstract

**Background:** Otitis media with effusion is a common problem that lacks an evidence-based nonsurgical treatment option. We assessed the clinical effectiveness of treatment with a nasal balloon device in a primary care setting.

**Methods:** We conducted an open, pragmatic randomized controlled trial set in 43 family practices in the United Kingdom. Children aged 4–11 years with a recent history of ear symptoms and otitis media with effusion in 1 or both ears, confirmed by tympanometry, were allocated to receive either autoinflation 3 times daily for 1–3 months plus usual care or usual care alone. Clearance of middle-ear fluid at 1 and 3 months was assessed by experts masked to allocation.

**Results:** Of 320 children enrolled, those receiving autoinflation were more likely than controls to have normal tympanograms at 1 month (47.3% [62/131] v. 35.6% [47/132]; adjusted relative risk [RR] 1.36, 95% confidence interval [CI] 0.99 to 1.88) and at 3 months (49.6% [62/125] v. 38.3% [46/120]; adjusted RR 1.37, 95% CI 1.03 to 1.83; number needed to treat = 9). Autoinflation produced greater improvements in ear-related quality of life (adjusted between-group difference in change from baseline in OMQ-14 [an ear-related measure of quality of life] score −0.42, 95% CI −0.63 to −0.22). Compliance was 89% at 1 month and 80% at 3 months. Adverse events were mild, infrequent and comparable between groups.

**Interpretation:** Autoinflation in children aged 4–11 years with otitis media with effusion is feasible in primary care and effective both in clearing effusions and improving symptoms and ear-related child and parent quality of life. Trial registration: ISRCTN, No. 55208702.

See also page [949](http://www.cmaj.ca/lookup/volpage/187/949) and [www.cmaj.ca/lookup/doi/10.1503/cmaj.150527](http://www.cmaj.ca/lookup/doi/10.1503/cmaj.150527)

Otitis media with effusion, also known as glue ear, is an accumulation of fluid in the middle ear, without symptoms or signs of an acute ear infection. It is often associated with viral infection.1–3 The prevalence rises to 46% in children aged 4–5 years,4 when hearing difficulty, other ear-related symptoms and broader developmental concerns often bring the condition to medical attention.3,5,6 Middle-ear fluid is associated with conductive hearing losses of about 15–45 dB HL.7 Resolution is clinically unpredictable,8–10 with about a third of cases showing recurrence.11 In the United Kingdom, about 200 000 children with the condition are seen annually in primary care.12,13 Research suggests some children seen in primary care are as badly affected as those seen in hospital.7,9,14,15 In the United States, there were 2.2 million diagnosed episodes in 2004, costing an estimated $4.0 billion.16 Rates of ventilation tube surgery show variability between countries,17–19 with a declining trend in the UK.20

Initial clinical management consists of reasonable temporizing or delay before considering surgery.13 Unfortunately, all available medical treatments for otitis media with effusion such as antibiotics, antihistamines, decongestants and intranasal steroids are ineffective and have unwanted effects, and therefore cannot be recommended.21–23 Not only are antibiotics ineffective, but resistance to them poses a major threat to public health.24,25 Although surgery is effective for a carefully selected minority,13,26,27 a simple low-cost, nonsurgical treatment option could benefit a much larger group of symptomatic children, with the purpose of addressing legitimate clinical concerns without incurring excessive delays.

Autoinflation using a nasal balloon device is a low-cost intervention with the potential to be used more widely in primary care, but current evidence of its effectiveness is limited to several small hospital-based trials28 that found a higher rate of tympanometric resolution of ear fluid at 1 month.29–31 Evidence of feasibility and effectiveness of autoinflation to inform wider clinical use is lacking.13,28 Thus we report here the findings of a large pragmatic trial of the clinical effectiveness of nasal balloon autoinflation in a spectrum of children with clinically confirmed otitis media with effusion identified from primary care.

## Methods

### Study design and participants

We carried out an open, pragmatic randomized controlled trial in primary care. We examined the difference in effectiveness between autoinflation 3 times daily for 1–3 months plus usual care and usual care alone.

The study was piloted in 4 practices. The main study recruited children from 43 general practices from 17 primary care trusts (independent local groups) in the UK, between January 2012 and February 2013. Most (89%) children were identified by practice-based computer search, and the rest were recruited through opportunistic case finding by practitioners, nurses and health visitors. Tympanometry and recruitment were undertaken by practice-based nurses. The study protocol is available in Appendix 1 at [www.cmaj.ca/lookup/suppl/doi:10.1503/cmaj.141608/-/DC1](http://www.cmaj.ca/lookup/suppl/doi:10.1503/cmaj.141608/-/DC1).

The National Research Ethics Service gave ethics approval for this study. The 17 participating primary care trusts gave National Health Service approval.

### Eligibility criteria

Children were eligible for inclusion if they were attending school and aged 4–11 years (deemed an age likely to be able to comply with autoinflation); had a history of hearing loss or other relevant ear-related problems in the previous 3 months; and had objective otoscopic and tympanometric confirmation of otitis media with effusion in at least 1 ear (i.e., had 1 or 2 type-B tympanograms using a modified Jerger classification) at the point of randomization (Table 1).32,33 Children were excluded if they had current clinical features of acute otitis media (e.g., ear pain, fever or otoscopic features of acute inflammation), recent or planned ear surgery, a known latex allergy or a recent nosebleed.

View this table:
[Table 1:](http://www.cmaj.ca/content/187/13/961/T1)

Table 1: 
Tympanometric classification*† (based on a modified Jerger classification32,33) and interpretation34

### Randomization and masking

An independent external agency provided a centralized Web-based computer-generated randomization system ([www.sealedenvelope.com](http://www.sealedenvelope.com)) for nurses to access while recruiting children. The Oxford Primary Care Clinical Trials Unit independently managed, coordinated, analyzed and checked the data validity. The randomization involved an algorithm with minimization based on 3 variables: age, sex and baseline severity (bilateral v. unilateral type-B tympanograms).9 Because of the nature of the intervention, use of placebo was not possible, and therefore nurses, children and families were not masked to treatment allocation.

### Procedures

All participating parents and children received information sheets, and parents gave written informed consent for screening to the research nurse. Children were invited to give written assent.

The simple autoinflation method involved inflating a purpose-manufactured balloon (Otovent) by blowing through each nostril into a connecting nozzle31 (Figure 1). Children receiving treatment were instructed by watching the nurse or parent demonstrate the procedure after stretching the balloon. The schedule involved inflating the balloon 3 times daily for an initial period of 1 month. Children still recording a type-B tympanogram in either ear at 1 month were advised to continue with autoinflation for a further 2 months. At the end of the study, affected children in the arm receiving usual care were offered a 1-month treatment pack.

![Figure 1:](http://www.cmaj.ca/https://www.cmaj.ca/content/cmaj/187/13/961/F1.medium.gif)

[Figure 1:](http://www.cmaj.ca/content/187/13/961/F1)

Figure 1: 
(A) Child demonstrating use of the Otovent device. Reproduced with permission. (B) Illustration of otitis media with effusion and the mechanism of autoinflation.

Image courtesy of A. Kestrel Medical Ltd. B. Émilie Lacharité

All nurses received training in the study methods, including tympanometry and interpretation (from an audiologist), updates on otoscopy and ways of maximizing study compliance. Hand-held calibrated MTP10 tympanometers (with printout facilities) were used.

### Study outcomes

We assessed outcomes at 1 and 3 months after randomization, during which time natural resolution effects would be expected to occur for some children.7,9

Our main outcome was the difference between groups in the proportion of children showing definite tympanometric resolution (i.e., normal middle-ear pressure, defined by a type-A or -C1 tympanogram) in at least 1 affected ear per child at 1 and 3 months. Intermediate type-C2 tympanograms showing negative pressure were considered insufficiently stringent for resolution of fluid (Table 1).32,34 We chose tympanometric outcomes because they allowed blinding, have been well validated previously,2,34 are regarded as a good choice for primary care studies13,35 and are used in meta-analysis.28 Two members of the trial team, trained in tympanometry and masked to allocation, independently reviewed anonymized tympanometry printouts. The expert interrater agreement was 89%, and disagreement was settled by a third independent audiologist. We found a Cohen κ of 0.7 for the level of agreement between nurses and masked experts.

Ear-related quality of life was measured at 3 months using the OMQ-14, an instrument developed by a process of refinement and iteration from large clinical and trial datasets of otitis media with effusion that optimized item mapping onto the Health Utilities Index.7,9,15,36 In addition, parents completed weekly diaries to record symptoms, adverse events and compliance. Days during which parents reported their child had hearing loss, earache, days off school, days requiring pain relief and sleep disturbance were summarized as days with any problem. We did not conduct pure tone audiometry because it cannot be done with adequate precision in nonspecialist settings.

### Sample size

A total sample of 295 children was required to provide 90% power (5% α) to detect a treatment effect (odds ratio [OR]) of 2.4, and allowing for 15% loss to follow-up at 1 month.28 The study was also powered to detect a clinically important difference of 0.3 of a standard deviation (SD) in total OMQ-14 score.

### Statistical analysis

We performed a modified intention-to-treat analysis, ignoring reported compliance but excluding children for whom no outcome measurement could be made. The relative effect of autoinflation on the primary outcome at 1 month and at 3 months was estimated using a generalized linear model for binary data with log-link function, and adjusted for baseline covariates (tympanometric baseline severity, age, sex and primary care trust). We conducted an ear-based analysis at 1 and 3 months using generalized estimating equations.

We compared change from baseline quality of life (standardized OMQ-14 scores) using a linear mixed-effects model. Data from weekly symptom diaries of days with symptoms related to otitis media with effusion were summarized according to categories of the number of days with symptoms. Groups were compared using an ordinal logistic regression model.

Analyses were conducted using SAS version 9.3 and Stata version 13.0 according to a prespecified analysis plan.

## Results

Between December 2011 and February 2013, 1235 children were screened for eligibility and 320 (26%) were randomly assigned to standard care alone or autoinflation plus standard care (Figure 2). The main reasons for ineligibility were that children lacked a type-B tympanogram (2 children from each arm were subsequently withdrawn because of this), were not currently attending school or reported a recent nosebleed. Ineligible children reported fewer symptoms associated with otitis media with effusion in the preceding 3 months, and had fewer consultations for otitis media in the previous 12 months.

![Figure 2:](http://www.cmaj.ca/https://www.cmaj.ca/content/cmaj/187/13/961/F2.medium.gif)

[Figure 2:](http://www.cmaj.ca/content/187/13/961/F2)

Figure 2: 
Enrolment, allocation and follow-up of patients. Note: ITT = intention-to-treat, OME = otitis media with effusion, R = randomization.

Baseline characteristics of children who were randomly assigned were balanced between the 2 groups (Table 2). Trial demographic data are comparable to national figures, but 33% of participating parents (v. 27% nationally) had a university or postgraduate degree.37

View this table:
[Table 2:](http://www.cmaj.ca/content/187/13/961/T2)

Table 2: 
Baseline characteristics of participants, by treatment group

### Main results

Retention was good, with 8.4% lost to follow-up at 1 month and 12.2% at 3 months. Uninterpretable tympanograms due to poor technique (leakage or low canal volume), and clinical problems (wax or perforation) were similar between groups, leaving 131 children in the autoinflation arm and 132 in the usual care arm in the modified intention-to-treat analysis.

Compared with standard care, children receiving autoinflation achieved tympanometric resolution more often at 1 month (adjusted relative risk [RR] 1.36, 95% confidence interval [CI] 0.99 to 1.88) and at 3 months (adjusted RR 1.37, 95% CI 1.03 to 1.83, number needed to treat [NNT] = 9) (Table 3). Sensitivity analyses using multiple imputation for missing data yielded similar but slightly smaller RRs that did not achieve statistical significance. Analyses for individual ears, adjusted for correlation between ears within child, showed that tympanometric resolution was significantly more likely with autoinflation at 1 month (adjusted RR 1.38, 95% CI 1.01 to 1.87) and at 3 months (adjusted RR 1.41, 95% CI 1.05 to 1.88), consistent with our per-child (primary) analysis (Table 3).

View this table:
[Table 3:](http://www.cmaj.ca/content/187/13/961/T3)

Table 3: 
Tympanometric resolution at 1 month and 3 months, by study group

### Subgroup analyses

We conducted prespecified subgroup analyses of effects of age (< 6.5 yr v. ≥ 6.5 yr), severity (1 v. 2 type-B ears at baseline), OMQ-14 standardized total score (< 0 v. ≥ 0) and sex on the primary outcome. In all cases we found no differences in treatment effects between subgroups. *P* values for the interaction term (treatment by subgroup) in the model ranged from 0.3 to 0.5 (Appendix 2, available at [www.cmaj.ca/lookup/suppl/doi:10.1503/cmaj.141608/-/DC1](http://www.cmaj.ca/lookup/suppl/doi:10.1503/cmaj.141608/-/DC1)).

### Ear-related quality of life and diary symptoms

At 3 months, the mean change from baseline in the standardized OMQ-14 total scores was greater in the autoinflation arm than in the control arm by −0.33 points (95% CI −0.59 to −0.07). The adjusted difference between groups was −0.42 points (95% CI −0.63 to −0.22) (Figure 3). This score difference represents a treatment effect size of 0.48 of an SD. The mean improvement in baseline score was −0.69 (0.84 SD) points at 3 months for the treatment arm. Effects were consistent across individual OMQ-14 items (Appendix 3, available at [www.cmaj.ca/lookup/suppl/doi:10.1503/cmaj.141608/-/DC1](http://www.cmaj.ca/lookup/suppl/doi:10.1503/cmaj.141608/-/DC1)). Overall, children in the autoinflation arm had fewer days with any symptom or problem than children in the control arm at 1 month (median [interquartile range] 8 [2–16] v. 9 [4–17] d; OR 0.66, 95% CI 0.41 to 1.05) and at 3 months (median [interquartile range] 14 [6–28] v. 22 [8–35] d; OR 0.58, 95% CI 0.37 to 0.90) (Table 3).

![Figure 3:](http://www.cmaj.ca/https://www.cmaj.ca/content/cmaj/187/13/961/F3.medium.gif)

[Figure 3:](http://www.cmaj.ca/content/187/13/961/F3)

Figure 3: 
Standardized total scores on the OMQ-14 (an ear-related measure of quality of life) at baseline and at 3 months. Lower scores represent improvement.

### Compliance

A total of 89% of parents reported using the device “most” or “all of the time” during the first month of treatment, consistent with the daily compliance sticker charts. This level of compliance appears to be maintained in those continuing treatment up to 3 months (80%).

### Adverse events

We found very little difference between study arms in the number of children with nosebleed (15% v. 14%), but there were more reported respiratory infections in the treatment group (15% v. 10% of children). Most of the respiratory infections were mild afebrile rhinorrhea. Eight children receiving autoinflation (compared with 2 receiving usual care) reported otalgia (Table 4). Five children in the treatment group and 4 in the control group had an episode of acute otitis media. Two children in the treatment arm were withdrawn: 1 was admitted to hospital with mild/early mastoiditis and made a full recovery, and a second was withdrawn due to otalgia.

View this table:
[Table 4:](http://www.cmaj.ca/content/187/13/961/T4)

Table 4: 
Adverse events, by study group

## Interpretation

In this study, we observed that autoinflation in young, school-aged children with otitis media with effusion is feasible in primary care and effective in clearing middle-ear effusions and improving symptoms and ear-related child and parent quality of life. Autoinflation is a simple, low-cost procedure that can be taught to young children in a primary care setting with a reasonable expectation of compliance. With an NNT of 9, it is a relatively noninvasive option that can add benefit by helping to fill the current gap between either doing nothing effective or referring for surgery.13 Wider use of this device has considerable potential to address the present lack of treatment options for most symptomatic children, and the frequency with which inappropriate antibiotics continue to be used to fill this gap.9,12,13,21,24,25 Because fluid in the ear does not completely clear in many instances even after 3 months, and with a tendency to recur, clinical vigilance with the option for surgery remains crucial to evidence-based management.

Best evidence suggests there are currently no proven nonsurgical interventions for glue ear. Parents often see temporizing strategies as causing unreasonable delay, and this can also lead to use of ineffective treatments, such as antibiotics. The most recent Cochrane Review of autoinflation28 highlighted the need for a large primary care study. The small hospital-based trials available29–31 for meta-analysis did not assess autoinflation in a primary care setting and lack both power and generalizability to the majority of affected children. Adding our data (and unpublished pilot, *n* = 20) to the meta-analysis more than doubles the available sample size with an estimated aggregate effect size of 1.61 (95% CI 1.26 to 2.06, *I*2 heterogeneity 0.0%) (Appendix 4, available at [www.cmaj.ca/lookup/suppl/doi:10.1503/cmaj.141608/-/DC1](http://www.cmaj.ca/lookup/suppl/doi:10.1503/cmaj.141608/-/DC1)).

For the child, parent and professional, the main issue is the impact caused by otitis media with effusion.38,39 From this consequential perspective, moderate improvement in the total OMQ-14 score is important and encouraging (e.g., in terms of reduced days with hearing difficulty over 3 mo) (Appendix 3). Published data on ear-related quality of life from other trials of otitis media with effusion are currently very sparse.9,15,26

### Limitations

The main limitation of this study is that the intervention cannot be blinded, and Hawthorne effects are possible.40 However, concealment issues are unlikely to affect tympanometric outcomes, because Web-based randomization was used for allocation, and all printouts were anonymized and assessed by experts who were unaware of allocation. Even if symptom and mapped quality of life (OMQ-14) scores were affected by performance bias, the effects observed would still be likely to reflect routine practice. The study population included children who were likely to be able to reliably perform autoinflation (i.e., age ≥ 4 yr). Although children of all ages frequently present to primary care, in the UK the most common age for referral for otitis media with effusion is between 3.5 and 8 years.7,9,26 This study does not address treatment for the youngest group of children affected. However, children as young as 3 years have been found to be able to use the device in hospital settings.29,31

### Conclusion

We have found use of autoinflation in young, school-aged children with otitis media with effusion to be feasible, safe and effective in clearing effusions, and in improving important ear symptoms, concerns and related quality of life over a 3-month watch-and-wait period. Autoinflation may not be suitable for all children, especially those under 4 years of age, and does require ongoing commitment to treatment. Further research is needed for very young children, and to inform prudent use across different health settings. The method has scope to be used in many symptomatic children, and is capable of producing better management and outcomes in primary health care systems. Wider use of nasal balloon autoinflation could address the present lack of treatment options for children with symptomatic otitis media with effusion.

See the following video online: Demonstration of nasal balloon autoinflation. [www.cmaj.ca/lookup/suppl/doi:10.1503/cmaj.141608/-/DC2](http://www.cmaj.ca/lookup/suppl/doi:10.1503/cmaj.141608/-/DC2)

## Acknowledgements

The authors thank the following: the Health Technology Assessment and National Institute for Health Research funders; the Oxford Primary Care Clinical Trials Unit, especially Sarah Benge for protocol development, logistical help and pilot work; Sharon Charnock of Starkey Laboratories Ltd. for audiological review and assistance with nurse training; Mark Haggard and Helen Spencer for use of the OMQ-14, with advice and supply of a scoring system; and Sven-Eric Stangerup, the inventor of Otovent, for advice on use. The authors also thank the regional primary care research networks and all participating practices and Kestrel Medical Ltd. for the trial supply of Otovent and providing the photograph in Figure 1. Finally, and most importantly, the authors thank all the children and families who took part in the study.

## Footnotes

*   *CMAJ* Podcasts: author interview at [soundcloud.com/cmajpodcasts/141608-res](http://soundcloud.com/cmajpodcasts/141608-res)

*   **Competing interests:** None declared.

*   This article has been peer reviewed.

*   **Contributors:** Ian Williamson conceived and designed the study, led protocol development, funding application, analysis and interpretation. Jane Vennik provided day-to-day management, coordinated recruitment, and contributed to the data collection, analysis and interpretation. Anthony Harnden contributed to protocol development, funding application and data interpretation. Merryn Voysey led the statistical analysis and interpretation of the study findings. Rafael Perera contributed to protocol development, funding application, statistical analysis and data interpretation. Sadie Kelly supervised the randomization process, and coordinated data collection, cleaning and validation. Guiqing Yao and James Raftery led the health economic analysis. David Mant contributed to protocol development, funding application and interpretation of findings. Paul Little contributed to protocol development, funding application and interpretation of findings. All of the authors contributed to drafting the article. Ian Williamson and Jane Vennik revised the article, with particular support from Merryn Voysey and Anthony Harnden. All of the authors gave final approval of the version to be published and agreed to act as guarantors of the work.

*   **Funding:** This project was funded by the Health Technology Assessment program of the National Institute for Health Research and will be published in full in *Health Technology Assessment* ([www.nets.nihr.ac.uk](http://www.nets.nihr.ac.uk)). The funders had no role in the design or conduct of the study, the analysis or interpretation of the results, or the decision to submit the manuscript for publication.

*   Accepted June 10, 2015.

## References

1.  Zielhuis GA, Rach GH, van den Broek P. Screening for otitis media with effusion in preschool children. Lancet 1989;1:311–4.
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=2563465&link_type=MED&atom=%2Fcmaj%2F187%2F13%2F961.atom) 
    
    [Web of Science](http://www.cmaj.ca/lookup/external-ref?access_num=A1989T206800012&link_type=ISI) 

2.  Stool SEBA, Berman S, Carney CJ, et al. Otitis media with effusion in children. Rockville (MD): Agency for Healthcare Research and Quality; 2013.
    
    

3.  Fry J, Sandler G. Common diseases. 5th ed. New York: Kluwer Academic Publishers; 1993.
    
    

4.  Williamson IG, Dunleavey J, Bain J, et al. The natural history of otitis media with effusion — a three-year study of the incidence and prevalence of abnormal tympanograms in four South West Hampshire infant and first schools. J Laryngol Otol 1994; 108: 930–4.
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=7829943&link_type=MED&atom=%2Fcmaj%2F187%2F13%2F961.atom) 
    
    [Web of Science](http://www.cmaj.ca/lookup/external-ref?access_num=A1994PU50000002&link_type=ISI) 

5.  Higson J, Haggard M. Parent versus professional views of the developmental impact of a multi-faceted condition at school age: otitis media with effusion (‘glue ear’). Br J Educ Psychol 2005;75:623–43.
    
    [CrossRef](http://www.cmaj.ca/lookup/external-ref?access_num=10.1348/000709905X41906&link_type=DOI) 
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=16318682&link_type=MED&atom=%2Fcmaj%2F187%2F13%2F961.atom) 

6.  Maw AR. Glue ear in childhood. London (UK): Mac Keith Press; 1995.
    
    

7.  Browning GG. Two-year outcome of ventilation tubes in a randomized controlled trial of persistent childhood otitis media with effusion. Clin Otolaryngol Allied Sci 2001;26:342–4.
    
    

8.  Browning GG. Watchful waiting in childhood otitis media with effusion. Clin Otolaryngol Allied Sci 2001;26:263–4.
    
    [CrossRef](http://www.cmaj.ca/lookup/external-ref?access_num=10.1046/j.1365-2273.2001.00471.x&link_type=DOI) 
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=11559333&link_type=MED&atom=%2Fcmaj%2F187%2F13%2F961.atom) 

9.  Williamson I, Benge S, Barton S, et al. Topical intranasal corticosteroids in 4–11 year old children with persistent bilateral otitis media with effusion in primary care: double blind randomised placebo controlled trial. BMJ 2009;339:b4984.
    
    [Abstract/FREE Full Text](http://www.cmaj.ca/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6MzoiYm1qIjtzOjU6InJlc2lkIjtzOjE3OiIzMzkvZGVjMTZfMS9iNDk4NCI7czo0OiJhdG9tIjtzOjIxOiIvY21hai8xODcvMTMvOTYxLmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

10. van Balen FA, de Melker RA, Touw-Otten FW. Double-blind randomised trial of co-amoxiclav versus placebo for persistent otitis media with effusion in general practice. Lancet 1996; 348:713–6.
    
    [CrossRef](http://www.cmaj.ca/lookup/external-ref?access_num=10.1016/S0140-6736(96)02511-1&link_type=DOI) 
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=8806290&link_type=MED&atom=%2Fcmaj%2F187%2F13%2F961.atom) 
    
    [Web of Science](http://www.cmaj.ca/lookup/external-ref?access_num=A1996VG37400010&link_type=ISI) 

11. Casselbrant ML, Brostoff LM, Cantekin EI, et al. Otitis media with effusion in preschool children. Laryngoscope 1985;95:428–36.
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=4039020&link_type=MED&atom=%2Fcmaj%2F187%2F13%2F961.atom) 
    
    [Web of Science](http://www.cmaj.ca/lookup/external-ref?access_num=A1985AFE4200011&link_type=ISI) 

12. Williamson I, Benge S, Mullee M, et al. Consultations for middle ear disease, antibiotic prescribing and risk factors for reattendance: a case-linked cohort study. Br J Gen Pract 2006;56:170–5.
    
    [Abstract/FREE Full Text](http://www.cmaj.ca/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NDoiYmpncCI7czo1OiJyZXNpZCI7czoxMDoiNTYvNTI0LzE3MCI7czo0OiJhdG9tIjtzOjIxOiIvY21hai8xODcvMTMvOTYxLmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

13. Surgical management of otitis media with effusion in children. London (UK): National Collaborating Centre for Women’s and Children’s Health; 2008.
    
    

14. Marchisio P, Grasso D. Using differences and similarities between centres for power and generality. Helsinki (Finland): ESPO Conference; May 27, 2010.
    
    

15. Dakin H, Petrou S, Haggard M, et al. Mapping analyses to estimate health utilities based on responses to the OM8-30 Otitis Media Questionnaire. Qual Life Res 2010;19:65–80.
    
    [CrossRef](http://www.cmaj.ca/lookup/external-ref?access_num=10.1007/s11136-009-9558-z&link_type=DOI) 
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=19941078&link_type=MED&atom=%2Fcmaj%2F187%2F13%2F961.atom) 
    
    [Web of Science](http://www.cmaj.ca/lookup/external-ref?access_num=000273560800008&link_type=ISI) 

16. American Academy of Family Physicians; American Academy of Otolaryngology-Head and Neck Surgery; American Academy of Pediatrics Subcommittee on Otitis Media With Effusion. Otitis media with effusion. Pediatrics 2004;113:1412–29.
    
    [Abstract/FREE Full Text](http://www.cmaj.ca/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6MTA6InBlZGlhdHJpY3MiO3M6NToicmVzaWQiO3M6MTA6IjExMy81LzE0MTIiO3M6NDoiYXRvbSI7czoyMToiL2NtYWovMTg3LzEzLzk2MS5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 

17. Cullen KA, Hall MJ, Golosinskiy A. Ambulatory surgery in the United States 2006. Hyattsville (MD): National Center for Health Statistics; 2009.
    
    

18. Desai SN, Kellner JD, Drummond D. Population-based, age-specific myringotomy with tympanostomy tube insertion rates in Calgary, Canada. Pediatr Infect Dis J 2002;21:348–50.
    
    [CrossRef](http://www.cmaj.ca/lookup/external-ref?access_num=10.1097/00006454-200204000-00018&link_type=DOI) 
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=12075770&link_type=MED&atom=%2Fcmaj%2F187%2F13%2F961.atom) 
    
    [Web of Science](http://www.cmaj.ca/lookup/external-ref?access_num=000175150500017&link_type=ISI) 

19. van Dongen TM, van der Heijden GJ, Freling HG, et al. Parent-reported otorrhea in children with tympanostomy tubes: incidence and predictors. PLoS ONE 2013;8:e69062.
    
    [CrossRef](http://www.cmaj.ca/lookup/external-ref?access_num=10.1371/journal.pone.0069062&link_type=DOI) 
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=23874870&link_type=MED&atom=%2Fcmaj%2F187%2F13%2F961.atom) 

20. Grommet insertion and adenoidectomy for otitis media with effusion. Dublin (Ireland): Health Information and Quality Authority; 2013.
    
    

21. van Zon A, van der Heijden GJ, van Dongen TM, et al. Antibiotics for otitis media with effusion in children. Cochrane Database Syst Rev 2012;9:CD009163.
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=22972136&link_type=MED&atom=%2Fcmaj%2F187%2F13%2F961.atom) 

22. Simpson SA, Lewis R, van der Voort J, et al. Oral or topical nasal steroids for hearing loss associated with otitis media with effusion in children. Cochrane Database Syst Rev 2011; CD001935.
    
    

23. Griffin G, Flynn CA. Antihistamines and/or decongestants for otitis media with effusion (OME) in children. Cochrane Database Syst Rev 2011;CD003423.
    
    

24. Davies S, Grant J, Catchpole M. The drugs don’t work: a global threat. New York: Penguin; 2013.
    
    

25. Antimicrobial resistance: global report on surveillance. Geneva (Switzerland): World Health Organization; 2014. Available: [www.who.int/drugresistance/documents/surveillancereport/en](http://www.who.int/drugresistance/documents/surveillancereport/en) (accessed 2014 May 6).
    
    

26. Browning GG, Rovers MM, Williamson I, et al. Grommets (ventilation tubes) for hearing loss associated with otitis media with effusion in children. Cochrane Database Syst Rev 2010; CD001801.
    
    

27. Williamson I. Otitis media with effusion. BMJ Clin Evidence Jan. 12, 2011 [Epub ahead of print]. Available: [www.clinicalevidence.com/x/systematic-review/0502/overview.html](http://www.clinicalevidence.com/x/systematic-review/0502/overview.html) (accessed 2014 May 6).
    
    

28. Perera R, Glasziou PP, Heneghan CJ, et al. Autoinflation for hearing loss associated with otitis media with effusion. Cochrane Database Syst Rev 2013;5:CD006285.
    
    [CrossRef](http://www.cmaj.ca/lookup/external-ref?access_num=10.1002/14651858.CD006285.pub2&link_type=DOI) 
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=23728660&link_type=MED&atom=%2Fcmaj%2F187%2F13%2F961.atom) 

29. Blanshard JD, Maw AR, Bawden R. Conservative treatment of otitis media with effusion by autoinflation of the middle ear. Clin Otolaryngol Allied Sci 1993;18:188–92.
    
    [CrossRef](http://www.cmaj.ca/lookup/external-ref?access_num=10.1111/j.1365-2273.1993.tb00827.x&link_type=DOI) 
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=8365006&link_type=MED&atom=%2Fcmaj%2F187%2F13%2F961.atom) 

30. Ercan I, Cakir B, Kayaoglu S, et al. Long term effect of autoinflation in the treatment of otitis media with effusion. KBB Forum 2005;4:166–70.
    
    

31. Stangerup SE, Sederberg-Olsen J, Balle V. Autoinflation as a treatment of secretory otitis media. A randomized controlled study. Arch Otolaryngol Head Neck Surg 1992;118:149–52.
    
    [CrossRef](http://www.cmaj.ca/lookup/external-ref?access_num=10.1001/archotol.1992.01880020041013&link_type=DOI) 
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=1540344&link_type=MED&atom=%2Fcmaj%2F187%2F13%2F961.atom) 
    
    [Web of Science](http://www.cmaj.ca/lookup/external-ref?access_num=A1992HC81000012&link_type=ISI) 

32. Jerger J. Clinical experience with impedance audiometry. Arch Otolaryngol 1970;92:311–24.
    
    [CrossRef](http://www.cmaj.ca/lookup/external-ref?access_num=10.1001/archotol.1970.04310040005002&link_type=DOI) 
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=5455571&link_type=MED&atom=%2Fcmaj%2F187%2F13%2F961.atom) 

33. Fiellau-Nikolajsen M. Epidemiology of secretory otitis media. A descriptive cohort study. Ann Otol Rhinol Laryngol 1983; 92: 172–7.
    
    [CrossRef](http://www.cmaj.ca/lookup/external-ref?access_num=10.1177/000348948309200215&link_type=DOI) 
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=6838108&link_type=MED&atom=%2Fcmaj%2F187%2F13%2F961.atom) 

34. Cantekin EI, Bluestone CD, Fria TJ, et al. Identification of otitis media with effusion in children. Ann Otol Rhinol Laryngol Suppl 1980;89:190–5.
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=6778304&link_type=MED&atom=%2Fcmaj%2F187%2F13%2F961.atom) 

35. Browning GC. Paediatric otolaryngolgy. Chapter 72. In: Scott-Brown’s otorhinolaryngology, head and neck surgery. Vol 1, pt 12. 7th ed. London (UK): Hodder Arnold; 2008.
    
    

36. Feeny D, Furlong W, Torrance GW, et al. Multiattribute and single-attribute utility functions for the health utilities index mark 3 system. Med Care 2002;40:113–28.
    
    [CrossRef](http://www.cmaj.ca/lookup/external-ref?access_num=10.1097/00005650-200202000-00006&link_type=DOI) 
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=11802084&link_type=MED&atom=%2Fcmaj%2F187%2F13%2F961.atom) 
    
    [Web of Science](http://www.cmaj.ca/lookup/external-ref?access_num=000173594900006&link_type=ISI) 

37. Qualifications and students, local authorities in the United Kingdom. London (UK): Office for National Statistics; 2011. Available: [www.ons.gov.uk](http://www.ons.gov.uk) (accessed 2014 Jan. 8).
    
    

38. Rosenfeld RM, Goldsmith AJ, Tetlus L, et al. Quality of life for children with otitis media. Arch Otolaryngol Head Neck Surg 1997;123:1049–54.
    
    [CrossRef](http://www.cmaj.ca/lookup/external-ref?access_num=10.1001/archotol.1997.01900100019002&link_type=DOI) 
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=9339979&link_type=MED&atom=%2Fcmaj%2F187%2F13%2F961.atom) 
    
    [Web of Science](http://www.cmaj.ca/lookup/external-ref?access_num=A1997YA80900002&link_type=ISI) 

39. Hearing and balance disorders. Achieving excellence in diagnosis and management. London (UK): Royal College of Physicians; 2007.
    
    

40. McCarney R, Warner J, Iliffe S, et al. The Hawthorne effect: a randomised, controlled trial. BMC Med Res Methodol 2007; 7: 30.
    
    [CrossRef](http://www.cmaj.ca/lookup/external-ref?access_num=10.1186/1471-2288-7-30&link_type=DOI) 
    
    [PubMed](http://www.cmaj.ca/lookup/external-ref?access_num=17608932&link_type=MED&atom=%2Fcmaj%2F187%2F13%2F961.atom)