International Journal of Pediatric Otorhinolaryngology 114 (2018) 44–50

Contents lists available at ScienceDirect

International Journal of Pediatric Otorhinolaryngology

journal homepage: www.elsevier.com/locate/ijporl

Antibiotic prescription patterns for management of acute otitis media in T

Lebanon
Ali Nasrallaha, Ali Bacharoucha, Fadia Jaafarb, Mariam Ayyasha, R. Alexander Blackwoodc,∗
a
University of Michigan Medical School, Ann Arbor, MI, USA
b
Hammoud Hospital University Medical Center, Saida, Lebanon
c
Department of Pediatrics and Communicable Diseases, University of Michigan Medical School, Ann Arbor, MI, USA

A R T I C LE I N FO A B S T R A C T

Keywords: Objectives: The high incidence of Acute Otitis Media (AOM) along with low antibiotic efficacy in the treatment of
Acute otitis media AOM is particularly favorable for the emergence of antimicrobial resistance. The promotion of more con-
Practice guideline adherence servative antibiotic prescription habits has become an important focus of governments and academic societies.
Lebanon Little is known about the awareness and use of AOM practice guidelines by physicians in the Middle East. Our
Antibiotic prescription patterns
aim is to characterize AOM management in Lebanon by using an anonymous survey instrument to uncover
potential disparities in treatment trends and evaluate differences in clinical guideline adherence patterns.
Methods: A total of 75 practicing physicians were anonymously surveyed in Beirut, Saida, Nabatieh, Bekaa and
Tripoli, Lebanon. The survey tool used was previously used in Amman, Jordan by our colleagues at the
University of Michigan. The survey we used assessed awareness of and adherence to practice guidelines by
prompting responses to hypothetical AOM cases. Differences in performance between various physician groups
were noted.
Results: Overall, physician participants answered 67% of the survey questions correctly. Trainees did better
overall in terms of AOM management (62% correct responses as compared to 48% in attending physicians,
p = 0.0175). Trainees also performed better in terms of their ability to manage cases of potential AOM in two-
year old children and their ability to choose the appropriate medications (79% correct response rate compared to
71% in attending physicians, p = 0.0278). Participants who reported guideline adherence most or all of the time
had a 67% correct response rate in regards to their ability to diagnosis AOM, compared to a 57% correct response
rate in those who reported adhering sometimes or rarely to the guidelines (p = 0.0489). In the cases requiring
antibiotic treatment for body temp of over 39C with/without otalgia, only 47–57% of participants identified the
appropriate antibiotic regimen.
Conclusion: There are areas of potential improvement in adherence to clinical guidelines in the management,
diagnosis, and treatment of AOM by Lebanese physicians. Conducting interventions among physicians to im-
prove awareness of clinical guidelines and current treatment recommendations in Lebanon will likely improve
adherence to guidelines, enhance clinical outcomes, and may help advance the fight against antimicrobial re-
sistance.

1. Introduction multidrug resistant strains of bacteria, antibiotic prescription strategies

Acute Otitis Media (AOM) is one of the most common types of in-
fections in pediatric populations throughout the world affecting up to
80% of children by the age of three [1]. In the United States alone, the
diagnosis and management of AOM accounts for more than 30 million
clinic visits annually, a large proportion of which results in the pre-
scription of empiric antibiotics [1,2]. With the rising prevalence of
have become a major concern for governments and academic societies
and an important focus of numerous global public health initiatives. In
an effort to combat antibacterial resistance, many countries utilize local
or national antibiotic resistance surveillance data to create specific
clinical practice guidelines for their physicians [3,4]. The promotion of
more conservative antibiotic prescription habits is especially important
in the management of AOM as the high incidence and low antibiotic

∗
Corresponding author. University of Michigan Medical School, 1500 East Medical Center Drive, D5101 Medical Professional Building, SPC 5718, Ann Arbor, MI,
48109–5718, USA.
E-mail addresses: rab@med.umich.edu, abacharo@umich.edu (R.A. Blackwood).

https://doi.org/10.1016/j.ijporl.2018.08.014
Received 14 April 2018; Received in revised form 12 August 2018; Accepted 13 August 2018
Available online 16 August 2018
0165-5876/ © 2018 Published by Elsevier B.V.
A. Nasrallah et al. International Journal of Pediatric Otorhinolaryngology 114 (2018) 44–50

efficacy of this condition is particularly favorable for the emergence of Table 1

drug resistance [3]. Demographic Characteristics of Participants. Demographic characteristics of
In 2004, the American Academy of Pediatrics (AAP) and the the 75 physicians that participated in the survey.
American Academy of Family Physicians (AAFP) outlined numerous Participants (n = 75)
AOM clinical scenarios that should lead to observation as initial man-
agement without empiric antibiotic prescription [5]. Based on these Gender Males 36 (48.0)
Females 39 (52.0)
recommendations, non-severe and unilateral cases of AOM would be
Specialty Family Medicine 7 (9.3)
managed initially with symptom relief with the understanding that Internal Medicine 6 (8.0)
rescue antimicrobial therapy would be given only if symptoms wor- Pediatrics 57 (76.0)
sened after two to three days of observation [5]. Despite this effort, Otolaryngology 4 (6.7)
Training Country Lebanon 53 (74.7)
however, the proportion of AOM visits in the United States that result in
Russia 8 (11.3)
antibiotic prescription is about 80% even as numerous studies reveal France 3 (4.2)
that empiric treatment does not lead to better outcomes or fewer Syria 2 (2.8)
complications as compared to more conservative management options Others 5 (7.0)
[2,3,6–8]. Adherence to clinical practice guidelines will vary depending Practice Location Beirut 49 (65.3)
Saida 16 (21.3)
on the healthcare setting, education of the provider, availability of re-
Nabatieh 8 (10.7)
sources, and the country of practice. Some of the factors that lead to Bekaa 1 (1.3)
higher rates of empiric antibiotic prescription include lack of familiarity Tripoli 1 (1.3)
with published guidelines, lack of appropriate incentives by the Current Level Physician 40 (53.3)
Trainee 35 (46.7)
healthcare system, greater requirement for follow-up visits with ob-
Health Sector Private 27 (36.0)
servation, uncertainty of the diagnosis, pressure from the parents, and Public 12 (16.0)
concerns about potential complications that may arise with more con- Both 36 (48.0)
servative management options [3,9,10]. Primarily Practice Setting Inpatient 16 (21.3)
A study published by Mustafa et al. in the Journal of Epidemiology Outpatient 2 (2.7)
Both 57 (76.0)
and Global Health reports that the incidence of AOM in children five
Year of Training Completion 2004 or Prior 22 (34.4)
years and younger in Turkey, Oman, and Saudi Arabia ranges from 99 After 2004 42 (65.6)
to 207 per 1000 person-years, suggesting that AOM causes a substantial Year of Training Completion 2013 or Prior 26 (40.6)
burden to public health in these three Middle Eastern countries [11]. After 2013 38 (59.4)
While some data exists on the incidence of AOM in the Middle East,
Data is n (%).
little is known about the awareness and use of clinical practice guide-
Denominators may vary as a result of missing data.
lines by healthcare providers in the region. Our aim is to further
characterize AOM practice patterns in the region by using an anon-
2.1. Statistical analysis
ymous survey instrument similar to the one used in Iran by Kamrava
et al. and in Jordan by Khreesha et al. [9,12]. By conducting this study
Analysis was performed using SAS 9.4. All continuous variables
in Lebanon, we can better uncover potential disparities in treatment
were evaluated for normality using the Shapiro-Wilk statistic, histo-
trends and evaluate differences in clinical guideline adherence patterns.
grams and Q-Q plots. Continuous variables were described as
This knowledge can increase awareness around clinical guidelines and
mean ± standard deviation and compared using t-test. Categorical
combat the emergence of multidrug resistant strains that would lead to
data were described as absolute numbers and percentages and com-
our inability to fight serious bacterial infections.
pared using the chi-squared test.

2. Methods 3. Results

This study was determined to be exempt from Institutional Review
Board (IRB) review at the University of Michigan in accordance with
federal regulations. From January to February 2018, an anonymous
survey was conducted to assess both clinicians' awareness and knowl-
edge of the internationally utilized AAP/AAFP practice guidelines and
to characterize the clinicians' AOM practice patterns in Lebanon. The
study participants included pediatricians, internal medicine practi-
tioners, family medicine practitioners, and otolaryngologists at various
public and private institutions in Beirut, Nabatieh, Saida, Bekaa, and
Tripoli, Lebanon. Responses were collected via an anonymous online
survey instrument and administration of the survey in person for most
of the physicians. After explaining the scope of the study, informed
consent was obtained from all study participants. The survey was pro-
vided to us by our colleagues at the University of Michigan who utilized
it in a similar study in Amman, Jordan. The questions aimed to char-
acterize the clinicians' AOM practice patterns through questions fo-
cused on demographics, knowledge and attitudes towards AOM
guidelines, as well as questions on hypothetical AOM cases. The first set
of cases assessed the clinicians' abilities to diagnose AOM, the second
set assessed the clinicians' abilities to properly select observation vs
medical treatment, and the third set of cases assessed the clinicians'
understanding of appropriate antibiotic therapy.
A total of 75 physicians participated in this study between
December 2017 and February 2018 at multiple hospitals in Lebanon.
Characteristics of study participants are summarized in Table 1. 76%
were pediatricians. 75% had their medical training in Lebanon, 65%
practiced in Beirut, and 53% were attending physicians as compared to
47% who were trainees. 66% had completed (or will have completed)
their training after 2004 while 59% had completed (or will have
completed) their training after 2013. 2004 and 2013 correspond to the
years in which the two most recent AAP/AAFP AOM guidelines have
been released. Table 2 highlights the participants' familiarity and ad-
herence to AOM practice guidelines. 81% of participants reported being
familiar with and had access to established AOM practice guidelines.
89% of participants who utilize the AOM practice guidelines reported
using international sources. 81% of participants reported adhering with
established guidelines when treating patients with AOM either most or
all of the time.
Table 3 highlights part A of the survey addressing the management
of AOM. 85% of participants knew all the situations for when to refer a
patient to ENT. 55% answered correctly for when to recommend tym-
panostomy tubes. Of all the diagnostic criteria used in the diagnosis of
AOM, 56% of participants use symptomatology, 60% use stringent
otoscopic criteria, 23% use pneumatic otoscopy and 1% use tympano-
metry. When comparing differences in the total correct responses for

45
A. Nasrallah et al. International Journal of Pediatric Otorhinolaryngology 114 (2018) 44–50

Survey Part B - Physicians Ability to Diagnosis AOM Based on Eight Hypothetical Cases. Part B of the survey is presented. This section assessed the ability of the physicians to diagnosis AOM based on eight
Table 2

13 (17.3)

13 (17.3)

13 (17.3)

36 (48.0)

12 (16.0)
Unsure

4 (5.3)

5 (6.8)

6 (8.0)
Familiarity & adherence to otitis media practice guidelines.
Participants
(n = 75)

* 37 (49.3)

* 54 (72.0)

* 19 (25.3)

* 41 (54.7)
13 (17.6)
Are you familiar with and have access to Yes 61 (81.3)

6 (8.0)

5 (6.7)

5 (6.7)
established Otitis Media practice No 3 (4.0)

No
guidelines? Sometimes 11 (14.7)
If you do utilize Otitis media practice Local 1 (1.3)

* 65 (86.7)

* 56 (75.7)

* 64 (85.3)

* 57 (76.0)
Is it AOM?
guidelines, what source do you use? Regional 2 (2.7)

25 (33.3)

20 (26.7)

22 (29.3)
8 (10.7)
National 8 (10.7)

hypothetical cases. Total number of clinicians who answered each question (percent of survey population) is presented. The correct answers in bold and are represented by an asterisk.
International 67 (89.3)

Yes
How often do you adhere with All of the Time 16 (21.3)
established guidelines when treating Most of the Time 45 (60.0)

Body temperature of 38.5 °C for the past 2 days Red and bulging TM.
patients with Acute Otitis Media Sometimes 12 (16.0)

Body temperature of 39.5 °C, Red TM on otoscopy without a fluid

A body temperature of (38.5 °C). Distinct redness of the tympanic
(AOM)? Rarely 2(2.7)
Never 0 (0.0)

Data is n (%).
Denominators may vary as a result of missing data.
Total % may sum to more than 100 due to option for multiple choices selection.

Table 3

Reddened TM. Effusion in the middle ear.

A reddened and bulging TM on otoscopy.

membrane (TM) Fluid in the middle ear.

Annular redness around TM onotoscopy.

Survey Part A – Otitis Media Management. Part A of the survey is presented.

An equivocal orthoscopic examination.

This section assessed participants on basic practices that pertain to AOM
management, including referral, diagnostic criteria, and appropriate re-
commendations. Total number of clinicians who answered each question

collection in the middle ear.

(percent of survey population) is presented. The correct answers in bold and are
represented by an asterisk.

Physical Examination

Red TM on otoscopy.
Participants
(n = 75)

At what point should a patient be Evidence of anatomic 0 (0.0)

referred to an otolaryngologist damage
(ENT)? Evidence of hearing loss 11 (14.7)
Evidence of language 11 (14.7)

A 5-month-old infant with severe restlessness and a history of body temperature of 39.5 °C in the recent 12 h.

A restless 18- month-old girl with body temperature of 39 °C from yesterday who touches her left ear during

A 32-year-old woman with a history of upper respiratory tract infection within the last week; she complains
delay
*All of the above *64 (85.3)
A 15-year-old girl with moderate otalgia, and a history of 39 °C body temperature in recent five days.

A 10-month-old boy with restlessness and body temperature of 38 °C since yesterday and a history of
None of above 0 (0.0)
Other 3 (4.0)
What are the diagnostic criteria Symptomatology 42 (56.0)
you utilize in the diagnosis of Stringent otoscopic 45 (60.0)
acute otitis media? criteria
Pneumatic otoscopy NA 17 (22.7)
Tympanometry 1 (1.3)
*All of the above *24 (32.0)
None of above 0 (0.0)
Other 0 (0.0)
When would you recommend Children with 3–4 12 (16.0)
tympanostomy tubes? episodes of AOM in 6
A 25-year-old man complaining of acute onset of severe otalgia.

months
A 3-month-old infant with severe restlessness from 2 days ago.

His mother reports that he has had diarrhea and rhinorrhea.

Children with at least 4 7 (9.3)

AOM episodes within 12
months
A 5 year old boy with acute onset of severe otalgia.

Children 6mo-12 years 20 (26.7)

with bilateral OME
3 + months with
documented hearing
difficulties
Every p-values obtained for each case was < 0.05.

*All of the above *41 (54.7)

None of above 0 (0.0)
In children with tympanostomy Yes 48 (64.0)
of dull pain in her right ear.

tubes, is there a requirement *No *18 (24.0)

for routine, prophylactic water Sometimes 9 (12.0)
touching his right ear.

precautions (i.e. as ear plugs,

headbands, or avoidance of
Medical History

swimming)?
examination.

*Refers to correct answer.

Data is n (%).
Total % may sum to more than 100 due to option for multiple choices selection.
*Refers to correct answer.
Every p-values obtained for the 4 questions in this table was < 0.05.
Data is n (%).
Case Number
Table 4

2
3

4
5

46
A. Nasrallah et al. International Journal of Pediatric Otorhinolaryngology 114 (2018) 44–50

this section of the survey, the management of AOM, there was no dif-
ference between genders, training country, practice location or ad-
herence to AOM practice guidelines as highlighted in Table 7. From a
training level standpoint, however, trainees' knowledge was at 62% as
compared to a lower level in attending physicians at 48% (p = 0.0175).
Also, participants who had their training after 2004 had 60% of ques-
tions correct as compared to 46% in those who had their training in
2004 or prior (p = 0.0289). Moreover, participants who had their
training after 2013 had 61% of questions correct as compared to 46% in
those who had their training in 2013 or prior (p = 0.0201).
Table 4 highlights part B of the survey addressing physicians' ability
to diagnosis AOM based on eight hypothetical cases. In the 5 years old,
15 years old and 25 years old cases, participants were able to diagnose
AOM correctly to a high degree of over 75%. 55% though diagnosed
AOM correctly in the case of a 32 year-old individual. For cases be-
tween 3 and 18 months old, participants diagnosed AOM over 70%
correctly in 2 of the 4 respective cases. In the remaining 2 cases, the
percentages correct were 49% and 25%. When comparing differences in
the total correct responses for this section of the survey, the ability to
diagnose AOM, there was no difference in percentages for correct re-
sponses when it came to training level, years since training completion,
training country or practice location as highlighted in Table 7. In terms
of gender however, 71% of women diagnosed AOM correctly which was
higher when compared to 60% of males (p = 0.0095). Additionally, in
terms of adherence to AOM practice guidelines, those who reported
guideline adherence for most or all of the time had 67% correct re-
sponse rate when compared to those who reported adhering sometimes
or rarely to the guidelines, at 57% correct response rate (p = 0.0489).
Table 5 highlights part C of the survey addressing antibiotic pre-
scription patterns where one of the tympanic membranes is red and
bulging on otoscopy based on six hypothetical cases. In each case, re-
spondents were asked to choose either immediate antibiotic treatment
or an initial observation period in which antibiotic prescription would
be withheld as recommended by AAP/AAFP guidelines. In cases of in-
fants under 6 months or between 6 months and 2 years of age, parti-
cipants assigned the correct plans to a high level with correct answers
percentages varying between 72 and 95%. In the case of a child older
than 2 years of age, participants assigned the correct management plan
to a high degree of 84% correct in the case of a fever being less than
39C. However, when the fever was over 39C, only 60% of participants
gave the appropriate management plan. When comparing differences in
the correct responses for this section of the survey, assigning the ap-
propriate plan, there was no difference in percentages for correct re-
sponses when it came to gender, training level, years since training
completion, training country, practice location or adherence to AOM
practice guidelines as highlighted in Table 7.
Table 6 highlights part D of the survey which addresses the ability to
manage cases of potential AOM in two-year old children and their
ability to choose the appropriate medication in cases where antibiotics
are indicated. Participants gave the appropriate treatment option to a
high level, over 75%, for all 4 cases regardless of body temperature and
otalgia presence or severity. When it came to appropriate treatment
option, 80% of participants gave the appropriate treatment plan for the
case of AOM with persistent body temperature greater than 39C and/or
otalgia despite 72 h of high dose amoxicillin. Otherwise, in the two
other cases requiring antibiotic treatment for body temp of over 39C
with/without otalgia, only 47–57% of participants identified the ap-
propriate antibiotic regimen. There were no significant differences in
responses to this section overall when gender, years since training
completion, training country or adherence to AOM practice guidelines
were taken into account as highlighted in Table 7. Trainees had a 79%
correct response rate to this section which was higher than attending
physicians at 71% (p = 0.0278). Also, those who practiced in Beirut
had a 78% correct response rate which was higher than those who
practice in other regions of Lebanon with 68% correct response rate
(p = 0.0081).
In terms of the overall survey responses, no significant differences in
correct response rate were noted when it came to training level, years
since training completion, training country, practice location or ad-
herence to AOM practice guidelines as shown in Table 7. Females
overall responded correctly more than males with 70% correct re-
sponses overall as compared to 64% in males (p = 0.0172). Comparing
responses to the same survey administered in Jordan, as highlighted in
Table 8, no significant differences in the percentages of correct answers
were noted between either country in any of the survey sections or the
survey overall.
4. Discussion
The findings of our study were supportive of our initial hypotheses
that the current trends for the management and treatment of AOM in
Lebanon would be similar to that of Jordan and that there would be
some differences relative to the guidelines in the United States. There
were many similarities between the results found in Lebanon and the
previous study done by our colleagues in Amman, Jordan [9]. We were
also unable to find any significance between all portions of our survey
assessment comparing Jordan to Lebanon, supporting our hypothesis
that there would be no difference between the diagnosis and manage-
ment of AOM in Lebanon relative to Jordan. Overall, Lebanese physi-
cians scored a 67.1% (SD 9.9) compared to a 69.2% (SD 11.7) in Jor-
danian physicians. The percent of total correct answers per survey
section only differed by 0.4–2.1%. This is not too surprising given that
these two countries are in close proximity and share similar cultures
and comparable socioeconomic conditions. These trends are likely not
unique to these two Middle Eastern countries and they may be found
throughout the Arab world, similar to how high rates of antimicrobial
resistant bacteria are found throughout various Arab countries [13]. For
example, many Arab countries share a high prevalence of extended-

Table 5
Survey Part C - Antibiotic Prescription Patterns Via 6 Hypothetical Cases. Part C of the survey is presented. This section assessed antibiotic prescription patterns
via 6 hypothetical cases where one of the tympanic membranes is red and bulging on otoscopy. In each case, respondents were asked to choose either immediate
antibiotic treatment or an initial observation period in which antibiotic prescription would be withheld as recommended by AAP/AAFP guidelines. Total number of
clinicians who answered each question (percent of survey population) is presented. The correct answers in bold and are represented by an asterisk.
Case Number Case Observation Antibiotic Treatment

1 An infant younger than 6 months with fever less than 39 °C 20 (26.7) * 55 (73.3)
2 An infant younger than 6 months with fever more than 39 °C 4 (5.3) * 71 (94.7)
3 A 6 monthto 2yearold child with fever less than 39 °C * 54 (72.0) 21 (28.0)
4 A 6month to 2yearold child with fever more than 39 °C 13 (17.3) * 62 (82.7)
5 A child older than 2 years with mild otalgia and fever less than 39 °C * 63 (84.0) 12 (16.0)
6 A child older than 2 years with mild otalgia and fever more than 39 °C 30 (40.0) * 45 (60.0)

Data is n (%).
*Refers to correct answer.
The p-values obtained for cases 1–5 were all < 0.05.

47
A. Nasrallah et al. International Journal of Pediatric Otorhinolaryngology 114 (2018) 44–50

Table 6
Survey Part D - Ability to Manage Cases of Potential AOM in Two Year Old Children & Ability to Choose the Appropriate Medication in Cases Where
Antibiotics are Indicated. Part D of the survey is presented. This section assessed the participants on their ability to manage cases of potential AOM in two year old
children as well as on their ability to choose the appropriate medication in cases where antibiotics are indicated. Total number of clinicians who answered each
question (percent of survey population) is presented. The correct answers in bold and are represented by an asterisk.
Case Number Case Treatment options Antibiotic regimen

1 AOM with body temperature of less than 39 °C a. Initiate antibiotic therapy at first visit 11 (14.7%)
*b. Observe the patient and postpone antibiotic therapy for
48–72hr. Only initiate antibiotics if symptoms do not
improve. 64 (85.3%)
2 AOM with body temperature of less than 39 °C and a. Initiate antibiotic therapy at first visit 18(24.0%)
mild otalgia *b. Observe the patient and postpone antibiotic therapy for
48–72hr. Only initiate antibiotics if symptoms do not
improve. 57 (76.0%)
3 AOM with body temperature of greater than 39 °C *a. Initiate antibiotic therapy at first visit 61 (81.3%) Low-dose Amoxicillin (40–50 mg/kg/
day) 2 (2.7%)
*High-dose Amoxicillin
(80–90 mg/kg/day) 43 (57.3%)
b. Observe the patient and postpone antibiotic therapy for Low-dose Co-Amoxiclav (40–50 mg/
48–72hr. Only initiate antibiotics if symptoms do not improve. 14 kg/day) 1 (1.3%)
(18.7%) High-dose Co-Amoxiclav (80–90 mg/
kg/day) 22 (29.3%)
Azithromycin 1 (1.3%)
4 AOM with body temperature of greater than 39 °C *a Initiate antibiotic therapy at first visit 73 (98.7%) Low-dose Amoxicillin (40–50 mg/kg/
and severe otalgia day) 1 (1.3%)
*High-dose Amoxicillin
(80–90 mg/kg/day) 35 (46.7%)
Low-dose Co-Amoxiclav (40–50 mg/
kg/day) 2 (2.7%)
High-dose Co-Amoxiclav (80–90 mg/
kg/day) 35 (46.7%)
b. Observe the patient and postpone antibiotic therapy for Azithromycin 1 (1.3%)
48–72hr. Only initiate antibiotics if symptoms do not improve. 1
(1.3%)
5 AOM with persistent body temperature greater than Low-dose Amoxicillin (40–50 mg/kg/
39 °C and/or otalgia despite 72 h of high dose day) 0 (0.0%)
amoxicillin High-dose Amoxicillin (80–90 mg/
kg/day) 2 (2.7%)
Low-dose Co-Amoxiclav (40–50 mg/
kg/day) 0 (0.0%)
*High-dose Co-Amoxiclav
(80–90 mg/kg/day) 60 (80.0%)
Azithromycin 7 (9.3%)

Data is n (%) or mean (standard deviation) unless otherwise specified.

Denominators may vary as a result of missing data.
Every p-values obtained for each case was < 0.05.

Table 7
Percentages of total correct answers per each survey section by demographic criteria.
Gender Training Level Year of Training Year of Training Training Country Practice Location Adherence to Guidelines
Completion Completion

Male Female Physician Trainee 2004 or After 2013 or After Lebanon Outside Beirut Outside Most/All Sometimes/
Prior 2004 Prior 2013 Lebanonb Beirut the time Rarely

Part Aa 54.6 54.7 48.3 61.9 45.5 60.3 46.2 61.4 56.0 51.5 58.5 47.4 55.7 50.0
( ± 21.3) ( ± 28.1) ( ± 21.3) ( ± 27.0) ( ± 19.4) ( ± 27.8) ( ± 21.2) ( ± 27.4) ( ± 26.8) ( ± 19.9) ( ± 25.9) ( ± 21.4) ( ± 24.9) ( ± 25.3)
P 0.9902 0.0175 0.0289 0.0201 0.4837 0.0666 0.4405
Part B 60.1 70.5 65.6 65.4 66.5 63.7 63.0 65.8 65.3 65.9 67.1 62.5 67.4 57.1
( ± 17.6) ( ± 16.3) ( ± 18.3) ( ± 17.2) ( ± 16.5) ( ± 17.9) ( ± 17.8) ( ± 17.1) ( ± 18.1) ( ± 16.9) ( ± 17.8) ( ± 17.3) ( ± 17.4) ( ± 16.8)
P 0.0095 0.9483 0.5455 0.5286 0.8982 0.2869 0.0489
Part C 77.3 78.2 79.2 76.2 80.3 75.8 80.1 75.4 77.4 78.8 75.2 82.7 77.3 79.8
( ± 17.4) ( ± 14.9) ( ± 12.9) ( ± 19.1) ( ± 14.2) ( ± 18.1) ( ± 14.9) ( ± 18.1) ( ± 15.7) ( ± 17.2) ( ± 17.2) ( ± 12.9) ( ± 17.5) ( ± 7.1)
P 0.8122 0.439 0.3143 0.2789 0.7281 0.0528 0.4096
Part D 71.4 78.0 71.1 79.2 76.0 78.2 75.3 78.9 76.8 70.1 78.4 68.1 74.0 78.6
( ± 14.1) ( ± 17.6) ( ± 17.3) ( ± 14.0) ( ± 13.5) (13.8) ( ± 12.5) (14.4) ( ± 16.6) (14.6) ( ± 16.3) ( ± 14.2) ( ± 15.6) ( ± 16.4)
P 0.0789 0.0278 0.5344 0.2945 0.1048 0.0081 0.3463
All Parts 64.3 69.7 65.7 68.8 67.3 67.7 66.0 68.6 67.7 65.8 68.5 64.6 67.5 65.4
( ± 9.1) ( ± 10.0) ( ± 9.9) ( ± 9.8) ( ± 10.9) ( ± 9.4) ( ± 10.6) ( ± 9.3) ( ± 10.4) ( ± 8.7) ( ± 10.7) ( ± 7.8) ( ± 10.1) ( ± 9.0)
P 0.0172 0.1784 0.8664 0.2977 0.4584 0.1077 0.4759

Data is mean (standard deviation).

a
The Total % Correct excluded answers for question 2: "What are the diagnostic criteria you utilize in the diagnosis of acute otitis media?" given that it was
opinion-based.
b
All other categories in either 'Training Country' or 'Practice Location' were grouped together if not Lebanon or Beirut, respectively.

48
A. Nasrallah et al.
Table 8
Percentages of total correct answers per each survey section by survey ad-
ministration location.
Survey Administration Location
Lebanon (n = 75) Jordan (n = 71)
Survey Part A 54.7 ( ± 24.9) 53.5 ( ± 26.1)
P 0.7766
Survey Part A 65.5 ( ± 17.7) 65.1 ( ± 15.5)
P 0.8850
Survey Part A 77.8 ( ± 16.1) 77.0 ( ± 17.0)
P 0.7707
Survey Part A 74.9 ( ± 16.2) 77.0 ( ± 22.3)
P 0.5145
All Survey Parts Combined 67.1 ( ± 9.9) 69.2 ( ± 11.7)
P 0.2428
Data is mean (standard deviation).
*Means and standard deviations for the Jordan group were calculated through
summing up the percentages correct for the respective survey parts and cal-
culating the standard deviations accordingly.
**The Total % Correct excluded answers for question 2: "What are the diag-
nostic criteria you utilize in the diagnosis of acute otitis media?" given that it
was opinion-based.
spectrum beta-lactamase (ESBL) and carbapenemase-producing gram-
negative bacilli [14]. Lack of adherence to practice guidelines with
respect to antibiotic usage in the Middle East could be a contributing
factor to these trends.
A major difference between Lebanese practices relative to the cur-
rent clinical guidelines was on the current methods utilized by Lebanese
physicians in the diagnosis of AOM. Most physicians used stringent
otoscopic criteria (60%) and symptomatology (56%) when diagnosing
AOM. However, only 32% made the diagnosis based on all four criteria
as recommended by AAP/AAFP: symptomatology, stringent otoscopic
criteria, pneumatic otoscopy, and tympanometry. This is likely due to
the resources available for the physician and the patient. It is a common
theme to do more with less in resource scarce populations. Although the
technology is available, the average Lebanese patient may not be able
to afford a comprehensive exam with pneumatic otoscopy and tympa-
nometry.
More effort should be taken by physicians to remain active in their
learning and to stay updated on the recommendations for AOM man-
agement and treatment. Resident physicians did significantly better
than attending physicians on the management (Part A) and treatment
(Part D) portions of the survey similarly to the Jordan study(1).
Residents scored a 61.9% on part A compared to the attending physi-
cians 48.3%, and a 79.2% on Part D compared to 71.1% by attending
physicians. This is likely due to residents training following the release
of the 2004 American Academy of Pediatrics and American Academy of
Family Physicians' AOM guidelines. A similar correlation was also ob-
served in physicians who completed their training following the joint
AAFP and AAP 2004 guidelines and the most recent AAP 2013 guide-
line. They did significantly better on the AOM management portion of
the survey (Part A).
When asked about their adherence to established guidelines for the
treatment of patients with AOM, 61% of physicians in Lebanon self-
reported adhering to guidelines most of time or all of the time, com-
pared to 52% of the physicians surveyed in the Jordan study [9].
However, physicians in Jordan were more likely to state they had ac-
cess to AOM guidelines (89.9% vs 81.3%), supporting the theory ori-
ginating by colleagues in the Jordan study that access to guideline re-
commendations may not be the sole reason for compliance to guidelines
[9]. Numerous other factors can play a role in limited adherence to
guidelines including community perceptions and pressure, concern for
medication compliance, easy accessibility of antibiotics, and concerns
for possible complications if a decision to withhold treatment was made
[9].
49
International Journal of Pediatric Otorhinolaryngology 114 (2018) 44–50
Nonadherence to guidelines, and the use of unnecessary antibiotics,
is concerning given the global rise of antibiotic resistance. Although
most physicians appropriately prescribed antibiotics in the clinical
scenarios, there was still a high percentage of overprescribing and
prescribing of broader spectrum antibiotics as a first line therapy. For
example, our data showed that 46.7% of physicians would give Co-
Amoxiclav as a first line therapy for treating AOM in a patient with
body temperature > 39C and severe otalgia, contrary to recent guide-
lines suggesting the use of Amoxicillin [15]. The prevalence of non-
adherence to practice guidelines is not limited to Lebanon as numerous
studies report relatively low AOM guideline adherence rates in the
United States [16,17]. A 2007 study by Vernacchio et al. revealed that
agreement with the recommended antibiotic choices of the 2004 AAP/
AAFP AOM guidelines in the United States ranged from 12.7% to
57.2%, depending on the clinical scenario [16].
When participants in this study were asked for comments, some
shared their fear that whether they prescribed antibiotics or not, many
patients would still gain access to antibiotics without the prescription of
the physician. Pharmacies in Lebanon often do not request proof of
prescriptions prior to dispensing medications. One study surveying
Lebanese pharmacists found that 32% of antibiotics were dispensed
without medical prescription [18]. This may lead to physicians pre-
scribing medications to please parents, limit the use of broader spec-
trum antibiotics, as well as be able to monitor a patient's progress on
current medications. Another common factor affecting adherence to
guidelines was whether physicians felt that parents would be compliant
to the medication regimen. Based on some comments received as part of
the study, numerous physicians pointed out that they were more likely
to prescribe IM Ceftriaxone if they felt that it would be hard for a family
to adhere to an antibiotic plan or return to the clinic for follow up.
It is also important to recognize that the region in which a physician
works can affect how they practice medicine. Our study found that
physicians practicing in Beirut did better in treating AOM in children
greater than two years of age. This is likely due to more readily avail-
able resources in Beirut, increased presence of academic institutions,
and higher literacy rates amongst the population [19]. Another study
found that dispensing injectable antibiotics without medical prescrip-
tion was significantly higher in lower socio-economic areas of Lebanon
[18].
This study has some important limitations that need to be con-
sidered. Most physicians that were surveyed were pediatricians. This
creates a difference when comparing the results to the Jordan study in
which most physicians were otolaryngologists. Another potential lim-
itation is language barriers. Although the study was available in both
English and Arabic, most study participants elected to utilize the
English version. Of note, there was room in the survey for feedback.
None of the participants indicated that language was a barrier to filling
out the survey. A third limitation of the study is the use of a non-vali-
dated survey, although the survey utilized was based on published work
that was created by an expert panel with the aim of formulating cases
that can assess AOM practices. A fourth limitation is reflected in our
correct answer by demographic distribution analysis. While this section
provided valuable insight on the differences in knowledge levels when
comparing it based on a specific demographic parameter distribution, it
did not control for confounding factors that could influence the result
distribution.
5. Conclusion
Lebanese physicians performed similarly to Jordanian physicians in
our assessment of current AOM clinical practices. Although most phy-
sicians did well on our survey, there are areas of improvement in re-
gards to adherence to clinical guidelines in the management, diagnosis
and treatment of AOM by Lebanese physicians. Conducting interven-
tions amongst physicians to improve awareness of clinical guidelines
and the current treatment recommendations in Lebanon will likely
A. Nasrallah et al.
improve adherence to guidelines, clinical outcomes and be a positive
step in the fight against antimicrobial resistance.
Declaration of interest
The authors report no conflicts of interest.
Acknowledgements
The study was supported by funding from the University of
Michigan Medical School GlobalREACH program. The authors would
like to acknowledge Dr. Brent Williams, the Director of The Global
Health and Disparities program at the University of Michigan, and Dr.
Mohamad R. Issa for their valuable input and assistance throughout the
project.
References
[1] J. Rosa-Olivares, A. Porro, M. Rodriguez-Varela, G. Riefkohl, I. Niroomand-Rad,
Otitis media: to treat, to refer, to do nothing: a review for the practitioner, Pediatr.
Rev. 36 (11) (2015 Nov) 480–486, https://doi.org/10.1542/pir.36-11-480 quiz
487-8.
[2] I. Williamson, S. Benge, M. Mullee, P. Little, Consultations for middle ear disease,
antibiotic prescribing and risk factors for reattendance: a case-linked cohort study,
Br. J. Gen. Pract. 56 (524) (2006 Mar) 170–175 PubMed PMID: 16536956; PubMed
Central PMCID: PMC1828259.
[3] M. Haggard, Poor adherence to antibiotic prescribing guidelines in acute otitis
media—obstacles, implications, and possible solutions, Eur. J. Pediatr. 170 (2011)
323 https://doi.org/10.1007/s00431-010-1286-4.
[4] World Health Organization's strategy to contain resistance to antimicrobial drugs,
Rev. Panam. Salud Públic 10 (4) (2001 Oct) 284–294 Spanish. PubMed PMID:
11715174.
[5] American Academy of Pediatrics Subcommittee on Management of Acute Otitis
Media, Diagnosis and management of acute otitis media, Pediatrics 113 (5) (2004
May) 1451–1465 Review. PubMed PMID: 15121972.
[6] S. Sanders, P.P. Glasziou, C.B. Del Mar, M.M. Rovers, Antibiotics for acute otitis
media in children, Cochrane Database Syst. Rev. (1) (2004) CD000219, , https://
doi.org/10.1002/14651858.CD000219.pub2.
[7] L. Koopman, A.W. Hoes, P.P. Glasziou, C.L. Appelman, P. Burke, D.P. McCormick,
R.A. Damoiseaux, N. Le Saux, M.M. Rovers, Antibiotic therapy to prevent the de-
velopment of asymptomatic middle ear effusion in children with acute otitis media.
A meta-analysis of individual patient data, Arch. Otolaryngol. Head Neck Surg. 134
50
International Journal of Pediatric Otorhinolaryngology 114 (2018) 44–50
(2) (2008) 128–132, https://doi.org/10.1001/archoto.2007.3.
[8] P. Little, C. Gould, M. Moore, G. Warner, J. Dunleavey, I. Williamson, Predictors of
poor outcome and benefits from antibiotics in children with acute otitis media:
pragmatic randomised trial, BMJ 325 (7354) (2002 Jul 6) 22 discussion 22.
PubMed PMID: 12098725; PubMed Central PMCID: PMC116668.
[9] L. Khreesha, A. Bacharouch, R.A. Blackwood, M. Alkhoujah, M.R. Issa, The use of
practice guidelines in the management of pediatric cases of Acute Otitis Media in
Amman, Jordan, Int. J. Pediatr. Otorhinolaryngol. 96 (2017 May) 39–46, https://
doi.org/10.1016/j.ijporl.2017.02.024 Epub. 2017. Feb 27. PubMed PMID:
28390611.
[10] M.D. Cabana, C.S. Rand, N.R. Powe, A.W. Wu, M.H. Wilson, P.A. Abboud,
H.R. Rubin, Why don't physicians follow clinical practice guidelines? A framework
for improvement, J. Am. Med. Assoc. 282 (15) (1999, Oct 20) 1458–1465 Review.
PubMed PMID: 10535437.
[11] G. Mustafa, A.Y. Aidaroos, I.S. Abaidani, Meszaros, et al., Incidence and economic
burden of acute otitis media in children aged up to 5 years in three Middle Eastern
countries and Pakistan: a multinational, retrospective, observational study, J.
Epidemiol. Glob. Health 7 (2) (2017) 123–130, https://doi.org/10.1016/j.jegh.
2016.12.004.
[12] K. Kamrava, M. Jalessi, A. Asghari, M. Farhadi, A. Ahmadvand, B. Ghalehbaghi,
M. Saffari Jourshari, M.A. Motiei Jouibari, Have guidelines affected ear, nose, and
throat specialists' diagnoses and the prescription of antibiotics for acute otitis
media? Iran J. Otorhinolaryngol. 24 (67) (2012, Spring) 55–62 PubMed PMID:
24303386; PubMed Central PMCID: PMC3846213.
[13] M. Farah, G. Araj, Z. Daoud, R. Moghnieh, P. Salameh, D. Saade, R. Husni,
Surveillance of antimicrobial resistance in Lebanese hospitals: retrospective na-
tionwide compiled data. Lebanese Society of Infectious Diseases Study Group (LSID
study group), Int. J. Infect. Dis. 46 (2016) 64–70.
[14] H.M. Zowawi, H.H. Balkhy, T.R. Walsh, D.L. Paterson, β-Lactamase production in
key gram-negative pathogen isolates from the arabian peninsula, Clin. Microbiol.
Rev. 26 (3) (2013) 361–380, https://doi.org/10.1128/CMR.00096-12.
[15] A.S. Lieberthal, A.E. Carroll, T. Chonmaitree, et al., The diagnosis and management
of acute otitis media, Pediatrics 131 (3) (2013) e964–e999.
[16] L. Vernacchio, R.M. Vezina, A.A. Mitchell, Management of acute otitis media by
primary care physicians: trends since the release of the 2004 American Academy of
Pediatrics/American Academy of Family Physicians clinical practice guideline,
Pediatrics 120 (2007) 281–287 https://doi.org/10.1542/peds.2006-3601.
[17] M.G. Stewart, S. Manolidis, R. Wynn, M. Bautista, Practice patterns versus practice
guidelines in pediatric otitis media, Otolaryngol. Head Neck Surg. 124 (2001)
489–495 https://doi.org/10.1067/mhn.2001.115497.
[18] R. Farah, N. Lahoud, P. Salameh, N. Saleh, Antibiotic dispensation by Lebanese
pharmacists: a comparison of higher and lower socio-economic levels, J. Infect.
Public Health 8 (1) (2015 Jan-Feb) 37–46, https://doi.org/10.1016/j.jiph.2014.07.
003 Epub 2014 Aug 16. PubMed PMID: 25139454.
[19] reportThe National Human Development Report: Towards a Citizen's State,
Lebanon 2008-2009, Web, http://hdr.undp.org/sites/default/files/nhdr_lebanon_
20082009_en.pdf.