OME and Speech Therapy

A new paper has been published (Brennan-Jones et al, JSLHR, 2020) that examines the relationship between the outcome of a single tympanostomy assessment at age 6 with PPVT test scores at age 6 and 10 and CELF scores at age 10. My doctoral thesis was on the topic of otitis media with effusion (OME) and I noticed that the authors curiously omitted the most important large sample prospective study from consideration in their introduction and their discussion. The omission seems to have been strategic. The authors were motivated to compare their own study to others that had been flawed by ascertainment bias. However, there are other excellent studies that used a prospective design with good quality sampling procedures. Furthermore, these other studies have the advantage of multiple assessments of middle ear function at an age that is of particular relevance to language development. It is instructive to consider the findings of the literature as a whole when attempting to draw conclusions about the clinical implications of Brennan-Jones et al findings. The study cannot stand alone. For this reason, I offer my own commentary.

  1. OME is Normal

The first and most important fact to understand about otitis media with effusion is that it is normal. Because it is a “silent” condition the fluid in the child’s middle ear can remain unnoticed for 30 or more days. Even worse, common treatments such as antibiotics or decongestants are quite useless when it comes to clearing up the fluid. Although infection is dangerous to the child’s health it is the fluid that impairs hearing and it is the fluid that is hardest to cure. So children can spend a lot of their life with suboptimal hearing. That study that Brennan-Jones et al ignored? It involved frequent prospective monitoring of middle-ear status in 2253 infants, from 61 days until 2 years of age (the Pittsburgh study by Paradise et al). The proportion of infants who were observed to have middle ear effusion more than once was 48%, 79% and 91% at ages 6, 12, and 24 months. On average these infants spent about 20% of their life with fluid in one or both ears. A similar study conducted in Boston (Teele, Klein & Rosner) followed children from birth to age 3 and recorded a range of 0 to 500 days with middle ear effusion and an average of 116 days. Half the sample experienced more than 90 days with middle ear effusion and almost half the sample had a bout of OME during their first year. To summarize, nearly every child gets at least one ear infection but the range of days with middle ear effusion varies greatly from child to child.

  1. OME Causes Significant Hearing Loss

Almost every paper that discusses the conductive hearing loss that is associated with OME describes it as “mild” because most children achieve pure tone average thresholds of 20 to 25 decibels during an episode of OME and only 10% suffer losses of greater than 40dB (see Roberts et al for review). However the amount of hearing loss changes greatly during each episode and greatly across the population of children who have OME. Furthermore, children require a much greater signal-to-noise ratio to achieve the same perceptual performance as an adult when identifying and discriminating speech signals. The same level of hearing loss that is mild for an adult with normal language abilities is significant for an infant or young child that is engaged with the task of learning his or her first language.

  1. OME is Associated with Variations in Language Development

Both the Pittsburgh study (2253 infants monitored prospectively from birth) and the Boston study (205 infants followed prospectively from birth) found that amount of time with middle ear effusion was correlated with language development. I am reproducing some of the date below, grouped according to days with MEE ascending down the rows and SES categories ascending across the columns. In both studies, MEE and SES are significant predictors of vocabulary knowledge. In the Pittsburgh study the vocabulary measure was parent report of productive vocabulary on the McArthur Communicative Development Inventory when the child was 24 months old. In the Boston sample, the measure of receptive vocabulary was the Peabody Picture Vocabulary Test.

  Pittsburgh Sample (Expressive Vocabulary)
  Low SES Mid SES High SES
Least MEE








Most MEE




  Boston Sample (Receptive Vocabulary)
  Low SES Mid SES High SES
Least MEE








Most MEE




How do we interpret these data? The first thing to notice is that variation in vocabulary size is normal (Fenson et al., 2000). At 24 months a child might produce no words or over 400. What accounts for this broad variation? It is common to call on genetic explanations but environmental inputs play a large role in vocabulary development specifically and SES and OME are both environmental variables. The point here is that OME does not cause language delay but it is one variable that helps to explain the large variation in early vocabulary development within the normal range.

  1. What are the clinical implications of the research on OME?

It is a rather common tactic to conclude that the research data indicating a correlation between OME and slightly slower growth in some aspect of language development (as reported in Brennan-Jones et al between age 6 and 10 years for example) is of no particular clinical interest. The reason for this conclusion is that the impact of OME is taken to be “small” because the mean test scores are all within the normal range. In other words, OME does not cause language impairment and therefore “no clinical implications.”

Let’s think about this from the perspective of an SLP treating one particular patient. I have in mind the most common type of patient treated by the pediatric SLP in the world (I can predict this from survey data and large scale caseload studies): a child aged somewhere between 4 and 7 with a mixed speech sound disorder and expressive language delay. We can expect an underlying impairment with phonological processing that has a heritable genetic cause (Bishop et al, 2008). The most important protective factor (Rvachew & Grawburg, 2006) will be the child’s vocabulary size—something that is highly malleable. If the child receives sufficient high-quality inputs, it will be a lot easier to bring phonological processing skills into the expected range and ensure acquisition of literacy skills. If the child has chronic OME, you don’t really care whether the OME has caused the child’s speech and language skills or not. Even though I would still argue that there is reason to be concerned about permanent effects of OME during the first year on the development of the auditory system, you can let the scientists worry about that. The issue is that this child cannot afford to lose a single word of language input. Because right now, intense high-quality language input is all we have in our treatment tool box. Let’s make sure that each child on our caseload can hear the precious minutes of therapy input that we are providing. And when we send them back to their noisy homes and classrooms with their homework books, let’s make sure they can participate in those activities to their maximum benefit. Hearing impairment affects everybody. And this child in particular doesn’t have any days to lose.



Leave a comment


  1. Aravind

     /  June 28, 2020

    Love your blogs….thank you for sharing your insights.

  2. Heather

     /  March 19, 2021

    This is a master class in critical thinking. Thank you.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: