Panel 5: Impact of Otitis Media on Quality of Life and Development

Abstract

Objective:

To summarize recent advances in knowledge on otitis media (OM) and quality of life (QoL) and development by synthesizing relevant research in this field published between June 1., 2015 until June 1., 2019.

Data Sources:

Systematic searches of PubMed, Embase and the Cochrane Library using predefined database-specific syntaxes.

Review Methods:

Articles selected were randomized controlled trials and observational studies with an adequate control group estimating treatment effects of OM including acute OM (AOM), recurrent AOM (RAOM), OM with effusion (OME), chronic OM (COM) and chronic suppurative OM (CSOM). Items included were Health Status, Health Status Indicators, Quality of Life, Functional Status, Specific Learning Disorder, Developmental Disabilities, Language Development Disorders, and Problem Behavior.

Results:

The electronic database searches yielded a total of 699 records. After screening titles and abstracts, we identified 34 potentially eligible articles. Of these, 18 were excluded. This left 15 articles suitable for inclusion.

Conclusions:

Although evidence is accumulating that OM may significantly impair children’s QoL and development as well as caregiver’s QoL, studies on this topic are relatively scarce and vary substantially in terms of methodological quality and outcome measurement instruments (OMI) used. In this review, studies have used 10 different OMIs capturing a wide range of OM symptoms as well as generic and disease-specific QoL outcomes. OM was associated with negative effects on auditory processing, language and speech development, school readiness, social competence, psychosocial wellbeing, and sleep. We found only four relevant randomized controlled trials, which mostly failed to demonstrate superiority of interventions in terms of QoL improvement and reports on reversibility are lacking. This underpins the urgent need for high quality studies in this field using validated and uniform OMIs. To facilitate interpretation and harmonization of study findings, we suggest and support the development of a core outcome set for the various OM entities that should include the most reliable and meaningful QoL and developmental OMIs.

Introduction:

WHO defines health as “state of complete physical, mental and social well-being and not merely the absence of disease” [1]. Health-related quality of life (QoL) is defined as a multifactorial construct covering four dimensions: physical complaints, mental condition, functional impairment in daily life and impairment of interpersonal relationships [2]. These dimensions are captured with specific items from the patient’s or caregiver’s perspective. This indicates that studies aimed at measuring treatment effects should include objective as well as subjective (patient reported) outcomes. From a patient perspective, improvement in subjective outcomes such as QoL, functional health status and development may be even more important than those observed in objective outcomes.

Since Facione et al. published the first study specifically aimed at measuring QoL in relation to otitis media (OM) in 1991 [3], an increasing number of studies that measure subjective outcomes in relation to OM have been published. Initially, studies focused on evaluating the negative impact of hearing loss caused by OM on the development of communication disorders and on school or academic performance. However, other factors affect daily life, such as chronic ear discharge, discomfort and disturbance of balance, disturbed sleep, and frequent doctor visits. The different nature of the spectra of OM diseases - ranging from acute OM (AOM), recurrent AOM (RAOM), OM with effusion (OME), chronic OM (COM) to chronic suppurative OM (CSOM) - has resulted in a variety of different outcome measure instruments (OMI) using an array of generic and disease specific instruments. Among these, the disease specific OM-6 questionnaire is by far the most applied instrument [4].

This state-of-the art review aims to present recent advances in the field of QoL in OM patients and their families since 2015 including the developmental effects of OM that affects individuals QoL.

Methods:

Data Sources and Searches

For this narrative review, systematic searches of PubMed, Embase and the Cochrane Library were performed from June 1, 2015 to June 1, 2019. The search strategy was otitis media (exploded) linked with relevant subheadings using database-specific syntaxes shown in Table 1.

Table 1.

Search strategy for otitis media and quality of life and development studies between June 1. 2015 and June 1. 2019

Pubmed	Embase	The Cochrane Library
((((((((((“Health Status”[Majr]) OR “Health Status Indicators”[Majr]) OR “Quality of Life”[Majr]) OR “functional status”) OR “Specific Learning Disorder”[Majr]) OR “Developmental Disabilities”[Majr]) OR “Language Development Disorders”[Majr]) OR “chronic otitis media”) OR “Problem Behavior”[Majr])) AND (((((“chronic suppurative otitis media”) OR “chronic otitis media”) OR “acute otitis media”) OR “Otitis Media with Effusion”[Majr]) OR “Otitis Media”[Majr])	(((otitis media)OR(secretory otitis media))AND((health status)OR(quality of life)OR(functional status)OR(disability evaluation)OR(functional assessment))) As in PubMed we searched on Major Topic (in Embase = focus) via Subject Heading	(((otitis media)OR(otitis media with effusion)) AND ((health status)OR(quality of life)OR(functional status)OR(disability evaluation)OR(functional health status)))

Study Selection

All review authors independently screened titles and abstracts of the records studies to assess potential relevance for inclusion (see Author Contribution). Any discrepancies were resolved by discussion in a meeting during the International Symposium on Recent Advances in Otitis Media in Los Angeles on June 9, 2019.

All studies related to the various entities of OM (ranging from AOM, RAOM, OME, COM to CSOM) on QoL and development were included irrespective of study design. We used the following items to extract relevant records: Health Status, Health Status Indicators, Quality of Life, Functional Status, Specific Learning Disorder, Developmental Disabilities, Language Development Disorders, and Problem Behavior. For studies aimed at estimating the effect of treatment on QoL and/or development in children or adults with OM, only randomized controlled trials (RCTs) with the patient as unit of randomization and observational studies with an adequate control group were included. Non-English language studies, animal studies and systematic reviews without relevant studies published after June 1., 2015 were excluded along with all observational studies without adequate control groups.

For each relevant study, we extracted the following data: study design, study population, aim, outcome measurement instruments and main findings. For RCTs, the Cochrane risk of bias tool was used to assess risk of bias. Guided by the Cochrane Handbook for Systematic Reviews of Interventions [4], we judged trials as high, low or unclear risk of bias. This was based on the following domains: sequence generation (selection bias), allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), selective outcome reporting (reporting bias), and other sources of bias. Non-randomized studies aimed at estimating treatment effects were judged as high risk of bias due to its inherent risk of bias due to confounding.

Results:

The electronic database searches yielded a total of 699 records. After screening titles and abstracts, we identified 34 potentially eligible articles. Of these, 18 were excluded: four were systematic reviews without relevant studies published after June 1, 2015 [6–9], one RCT used ears (instead of patients) as unit of randomization [10], and 13 were observational studies without adequate control group [11–23]. This left 15 articles suitable for inclusion [24–38].

The OMIs used in the included studies varied substantially (Table 2). These instruments capture a wide range of OM symptoms like hearing loss, sleep and pain as well as generic and disease-specific QoL outcome effects such as cognitive or academic performance, speech impairment, physical suffering, emotional distress, activity limitations, daily disturbances and caregiver concerns. This reflects that QoL is a highly subjective measure related to the individual preferences and underlines why validation of the instruments is of extreme importance.

Table 2.

Quality of Life (QoL) instruments used by the reviewed studies including validation status and outcome measures.

QoL Instrument	Type	Characteristics	Comments
Otitis Media 6 (OM-6) [4]	OM Specific	6 item (physical suffering, hearing loss, speech impairment, emotional distress, activity limitations, and caregiver concerns. Validated.	Measures child QoL related to OM, caregiver proxy, with higher scores indicate higher burden of disease and more impact of OM
Parental Acute Otitis Media Quality of Life (PAR-AOM-QoL) [24]	OM Specific	15 items, 2 scales: parent’s emotional disturbance (ES) and daily disturbance (DDS), with each scale scored from 0 (best) to 100 (worst). Validated	Measures parent QoL related to child’s OM.
Child Health Questionnaire Parent Form (CHQ-PF50) [42]	Generic	14 specific physical and psychosocial domains. Scores 0 to 100, with higher scores indicating better functioning and greater well-being. Validated	Parent proxy, Child QoL age 5–18 years
QoL disruption scale [33]	OM Specific	Disruption to family QoL rated from 0 (not at all) to 4 (extremely) with each otorrhea episode after VT placed, averaged. Not validated	Measures family QoL from otorrhea episodes.
Pediatric Quality of Life Inventory (PedsQL) [43]	Generic	4 domains: physical, emotional, social, school. Scores 0–100, 100 = best. Validated	Child QoL, with both child and parent-proxy forms
Health Utilities Index Mark 3 (HUI3) [44]	Generic	Health utilities, family of multi-attribute preference-based utility measures, with scores ranging from –0·36 to 1·00 (where higher scores indicate better health-related quality of life). Validated.	Both child and adult QoL
OM8–30 [45]	OM Specific	32 items, 6 facets of global health-ear problems-sleep patterns (GLEA), reported hearing difficulty (RHD), respiratory symptoms (RESP), behaviour problems (BEH), combined speech, language and educational prospects (SPEC) and parent quality of life (PQoL). Lower score = better QoL. Validated.	Both child and adult OM-specific QoL
Chronic Otitis Media Outcome Test-15 (COMOT-15) [46]	Chronic OM specific	15 items, 3 subscales: ear symptoms (ES), hearing function (HF), mental health (MH). 1 question each for general evaluation of QoL and frequency of doctor visits. Total score and subscores transformed to a 0–100 scale (percentage) Validated in part.	Patient QoL
Chronic Ear Survey [47]	Chronic OM specific	13-item survey, 3 subscales: activity restriction (AR), symptoms (S), medical resources (MR) Validated in part.	Patient QoL
EuroQol five-dimension questionnaire (EQ-5D) [48]	Generic	5 domains of mobility, self-care, usual activity, pain/discomfort, and anxiety/depression. Validated.	Self-report (adult & youth), interviewer, and proxy forms for QoL

Discussion:

Impact of OM on QoL and development

Three studies investigated the impact of AOM on QoL (Table 3) [24–26]. These studies used the OM-6, PAR-AOM-QoL, and EQ-5D questionnaires (Table 2). The studies showed that AOM negatively affected children’s and parental QoL, but the level of disturbance varied across studies. Parental QoL was affected more if parents perceived their child’s AOM episode as more severe, if AOM occurred in younger children and in those with RAOM.

Table 3.

Studies on impact acute otitis media and recurrent acute otitis media on quality of life

Author, Year	Country	Study design	Participants, Setting	Study Period	Study Aim	QoL Outcome (range)	Results and Conclusions
AOM
Crawford, 201724	Malaysia	Cohort	110 children, <5 y, AOM, and their caregivers ENT outpatient and other hospital departments (paediatrics, ER, general outpatient clinic)	March 2013 – April 2014	To assess the impact of AOM on both patient and caregiver QoL	Children: OM-6 (6 domains; adapted to 0–100) Caregivers: PAR-AOM-QOL (2 domains, ES and DDS, 0–100) Malaysian EQ-5D (0.131–1.00)	Children: mean OM-6: 24.70 (SD 14.03); greatest impact on child’s QoL domain: mean 66.55 (SD 20.25) Caregivers: mean ES: 24.66 (SD 17.14) mean DDS: 15.45 (SD 16.93) mean Malaysian EQ-5D: 0.92 (SD 0.10) Conclusions: AOM negatively affected parental QoL, but the level of disturbance was low. The overall OM-6 score was low with greatest impact of AOM on child’s QoL subdomain.
Holl, 201525	Germany, Italy, Spain, Sweden and UK	Cohort	852 chidren, 1–6 y, AOM (1,063 episodes)	July 2008–Jan 2009	To assess item-convergent and item-discriminant validity criteria and internal consistency reliability of AOM-specific parental QoL questionnaire To assess impact of AOM on parental QoL	AOM-specific parental QoL questionnaire (15 items, 1–5)	Questionnaire showed good to excellent internal consistency reliability for various items (Cronbach’s a 0.82–0.97). Score > 3: - interrupted sleep: 68.4% - worry: 51.0% - altered daily schedule: 44.6% - less leisure time: 41.5% Factors negatively impacting QoL: increased parental perception of AOM severity, young child age, and multiple AOM episodes. Conclusions: The AOM-specific parental QoL questionnaire showed good performance. Parental QoL was affected by AOM proportionally to severity, number of episodes and younger child age.
Indius, 201526	Denmark	Cross-sectional	342 children, mean 21.1mo, with and without parent-reported AOM (ear pain with fever) within 4w Nursery day-care	Feb–March 2018	To investigate the difference in QoL between children with and without symptoms of ear disease in last 4w	OM-6 (6 items, adjusted to scale of 0 to 100)	Mean OM-6 scores: - 4w group: 31.4 (SD 19.7) - non-4w group: 4.7 (SD 8.2) Difference in mean OM6-scores: 26.6 (95% CI 23.0 to 30.3) Conclusions: Children with a parent-reported AOM episode in previous 4 weeks had lower QoL score than those without a parental report of AOM in the last 4 weeks.
RAOM
Van Brink, 201927	Sweden	Cross-sectional	3835 children scheduled for VT in Swedish quality register Secondary and tertiary care	2010–2016	To evaluate pre-operative QoL in children scheduled for VT insertion	Generic questions related to QoL (’how does the ear diseas affect the child’s general wellbeing, 0–2)	OME vs RAOM (logistic regression analysis, 0 ref category):  - OR for 1 (a little): 0.54 (95% CI: 0.41–0.72)  - OR for 2 ( a lot): 0.33 (0.24–0.43) Conclusions: Parents of children with ROAM were more likely to find that the ear disease affected the child–s general wellbeing than parents of children with OME.
Kujala, 201728	Finland	Case-control;cases derived from RCT	149 children, 10–24 mo, RAOM with or without surgery Tertiary care Aged-matched controls from general child population	March 2002–June 2004 (RCT data)	To evaluate QoL in children with RAOM aged less than 24 mo	CHQ-PF50 (14 domains, 0–100)	At baseline, QoL of children with RAOM was significantly poorer in the majority of the CHQ-PF50 domains compared to children without AOM or those with 1–4 AOM episodes. At one-year follow-up, QoL of children with RAOM was significantly poorer in most of the CHQ-PF50 domains compared to children without AOM. Conclusions: RAOM negatively impacted the child’s QoL,
Heidemann, 201529	Denmark	Cohort	491 children, 6mo-6y, OME, RAOM, RAOM and OME receiving VT 71 with RAOM, 187 with RAOM and OME Secondary care	Feb 2011–March 2012	To investigate differences in QoL between diagnostic subgroups of children with OM	OM-6 (6 items, adjusted to scale of 0 to 100)	Baseline mean OM-6 scores: - RAOM (RAOM and RAOM/OME): 49.0 (SD 17.7) - OME: 39.3 (SD 18.7) Difference in mean OM6-scores: 9.7 (95% CI 6.5 to 13.0) Conclusions: Children with RAOM (RAOM and RAOM/OME combined) had a significantly lower QoL score than those with OME.

AOM: acute otitis media; CHQ-PF50: Child Health Questionnaire Parent Form 50; CI: confidence interval; DDS: daily disturbance; ENT: ear, nose, and throat; ER: emergency room; ES: emotional disturbance; EQ-5D: EuroQol five-dimension questionnaire; mo: months; OM-6: Otitis Media-6 questionnaire; OME: otitis media with effusion; OR: odds ratio; PAR-AOM-QOL: Parental Acute Otitis Media Quality Of Life questionnaire; QoL: Quality of life; RAOM: recurrent acute otitis media; SD: standard deviation; UK: United Kingdom; VT: ventilation tubes; y: years;

Three studies investigated the impact of RAOM on QoL (Table 3) [27–29]. RAOM had a negative impact on children’s QoL with parents of children with RAOM reporting a greater impact on QoL judged as frequent ear pain, fever, sleepless nights and time lost from school or work than parents of children with OME and those without ear disease.

The impact of OME on auditory processing was studied in 12 children (Table 4) [30]. This study found that the fluctuating hearing loss that is often associated with OME may alter auditory processing and interfere with language and speech development. Hence, focus should also be put on stimulating speech and language in the management of OME.

Table 4.

Studies on impact otitis media with effusion on quality of life

Author, Year	Country	Study design	Participants, Setting	Study Period	Study Aim	QoL Outcome (range)	Results and Conclusions
OME, COM, RAOM
Khavarghazala ni, 201630	Iran	Case control	12 children, 8–10y, OME	2008 to 2011	To investigate differences in auditory processing between children with a history of OME and normal controls	Dichotic digit test (DDT) - measures central auditory binaural integration skills Gap in noise (GIN) - measures ability to detect temporal gaps of silence	Significant difference in both outcomes between groups, GIN (p<0.001) and DDT (p=0.002). Conclusions: These findings support the hypothesis that fluctuating hearing loss may affect central auditory processing during critical periods.
Bell, 201631	Australia	Case control	Children with COM including OME/RAOM	2009	To investigate differences in scores between children with COM and normal controls.	Development data from the Australian Early Development Census (AEDC). Questions based on the Canadian early development Instrument. Domains: -Physical health and wellbeing -Social competence -Emotional maturity -Language and cognitive skills -Communication and general knowledge	OM children at risk of developmental vulnerability at school entry. Developmentally vulnerable/at risk on the AEDC (OM children) Domain (OR, CI, p-value): Physical wellbeing (1.32, 1.20–1.45, .001) Social competence (1.34, 1.22–1.47, .001) Emotional maturity (1.32, 1.20–1.44, .001) Communication skills and general knowledge (1.26, 1.14–1.38, .001) Language and cognitive skills (1.17, 1.07–1.27, .001) Conclusions: Children with OM have increased risk of poor school readiness even without having other comorbidities.
Chando, 201632	Australia	Systematic review	Parental views on OM including OME/RAOM, 17 qualitative studies, 284 participants	July 2015 (date latest search)	To synthesize parental views on OM	Domains: -diminishing competency -disrupting life schedules -social isolation -threatening normal development -taking ownership -valuing support -cherishing health	Main findings: The additional medical responsibilities and anxieties of parents caring for a child with OM, often discounted by clinicians, can be disempowering and disruptive. Chronicity can raise doubt about treatment efficacy and parental competency, and fears regarding their child’s development.

A large study involving more than two thousand children investigated the relationship between different pediatric chronic illnesses and school readiness [31]. Children with OM had increased risk of poor school readiness even without having other comorbidities (Table 4). This suggests that it could be helpful to also include OM when reviewing and evaluating health conditions eligible for additional support at school entry.

The importance of parental QoL in relation to chronic illness in children has until recently been overlooked. A recent systematic review synthesized 17 qualitative studies on parental views on OM (mainly OME) (Table 4) [32]. This review found that OM may impose a significant burden on the child and for periods of time even resemble a chronic disease. Many parents will have to rearrange their lives in order to best accommodate treatment and treatment failure as well as re-occurrence of disease may raise doubt on both treatment efficacy but also on their own competency as parents. The review authors emphasized the importance of clinicians focusing also on the concerns and psychosocial needs of the parents when managing the child with OM such as e.g., concerns regarding the child’s development, absence of work, repeated antibiotic treatments, and surgical interventions.

Generally, there has been growing attention to the effects of hearing loss (HL) on QoL. The perceived handicap from HL varies depending on the attitudes and perceptions of the individuals, and not just on the severity of the HL. Up to now, the effect of HL and hearing restoration on QoL has primarily been evaluated in patients with sensorineural HL. However, in patients with OM, QoL is not only determined by functional limitations on hearing, which can lead to communication problems and disturbances in social interaction and professional life, but also by symptoms such as persistent discharge from the ear or pain and disruptions to daily life from frequent doctor visits.

In OM research, many studies focus on hearing function or QoL measurement. Although all disease-specific questionnaires include a subscore for “hearing impairment” in OM, there are only a few studies that evaluate the extent of hearing impairment and QoL simultaneously. Two studies found a moderate to high correlation between postoperative hearing level and QoL after tympanoplasty and ventilation tube insertion (Table 5) [27,33].

Table 5.

Studies on impact of hearing loss on quality of life

Author, Year	Country	Study design	Participants, Setting	Study Period	Study Aim	QoL Outcome (range)	Results and Conclusions
OM
van Brink, 201927	Sweden	Retrospective review of national register data	3835 children with OM who underwent grommet insertion, mean age 4.4 yrs	2010–2016	To analyze the QoL aspect in the Swedish grommet register	4 QoL questions including QoL hearing score Postoperative pure tone average	Significant correlation between QoL hearing score and postoperative pure tone average (p<0.001). Conclusions: The degree of postoperative improvement in pure tone average correlated with the improvement in QoL.
Lucidi, 2019)33	Italy	Case series	81 adults, 18–70y, cholesteatomatous OM who underwent tympanomastoidectomy (CWD n=50, CWU n=31)	2011–2014	To assess whether outcomes vary between patients who underwent CWD and CWD tympanomastoidectomy and analyze the correlation between postoperative pure tone average and the COMOT-15 "Hearing Function" subsection scores	Chronic Ear Survey COMOT-15 Postoperative pure tone average	Significant association between postoperative pure tone average and subjective evaluation of hearing loss in QoL measurement (r = 0.721). Conclusions: The degree of postoperative improvement in pure tone average correlated with the subjective evaluation of hearing loss in QoL measurement .

Studies using multivariable analyses of different influencing factors such as hearing loss, psychological health and well-being, middle ear pathology, symptoms and intervention, socioeconomic status on QoL have not been published to date.

Impact of treatment for OM on QoL and development

Observational studies cannot provide definite evidence on any treatment effect due to its inherent risk of bias. This evidence should come from well-conducted RCTs.

We identified only four relevant RCTs (Table 6) [34–37]. Oral steroids did not have a significant effect on QoL and hearing in children with persistent OME and bilateral hearing loss, but balloon autoinflation may have in those with a recent onset of ear symptoms and OME in one or both ears. However, balloon autoinflation may not be suitable for children under the age of four years and requires ongoing commitment. Treatment of early tympanostomy tube otorrhea with antibiotic or saline ear drops had no differential impact on children’s QoL. Cartilage tympanoplasty combined with Eustachian tube balloon dilatation in the treatment of adhesive otitis media might improve QoL, although this evidence came from a trial with a high risk of bias (Table 6).

Table 6.

Randomized controlled studies on impact of treatment for OM on QoL and development

Author, Year	Country	Study design	Participants, Setting	Intervention (participants)	Comparator (participants)	QoL Outcome (range)	Risk of Bias	Results and Conclusions
Early tympanostomy tube otorrhea
Gabarain, 201934	USA	RCT	200 children, mean age 3.1y, VT insertion for RAOM/OME	Ciprofloxacin ear drops intraoperatively and postoperatively for 5 days (unknown)	Saline ear drops intraoperatively and postoperatively for 5 days (unknown)	QoL disruption scale (0–4)	High^	128 children (67 saline, 61 ciprofloxacine) were analyzed; Otorrhea disruption index (cipro vs saline): - 1w: 0.3 vs 0.4 - 2w: 1.7 vs 1.0 - 3w: 1.0 vs 0.6 −4w: 0.5 vs 1.0 - overall: 0.7 vs 0.7 Conclusions: Early tympanostomy tube otorrhea episodes did not heavily affect patients’ or families’ QoL. QoL disruption caused by these early otorrhea episodes was not significantly different between children who received saline or ciprofloxacin ear drops intra- and postoperatively.
OME
Francis, 201835	England and Wales	RCT	389 children, 2–8y, OME for ≥3 mo plus bilateral hearing loss ENT outpatient or audiology and audiovestibular clinics	Oral steroids1 , 7d (200)	Placebo (189)	OM8–30 PedsQL (0 to 100) HUI3 (−0.36 to 1.00) Acceptable hearing Reported difficulty hearing	Low	Adjusted between-group difference in means averaged across all follow-up timepoints (baseline, 5w,6mo,12mo): OM8–30: 0.05 (−0.12 to 0.22) PedsQL: 1.23 (0.66 to 2.27) HUI3: −85.11 (−420.65 to 50.44)* Acceptable hearing in 40% of placebo versus 33% of steroid group: absolute risk difference 7% (95% CI –3 to 17) Difference in reported difficulty hearing: 0.03 (– 0.13 to 0.20) Conclusions: Oral steroids has no significant effect on QoL and hearing as compared with placebo in children with persistent OME and bilateral hearing loss.
Williamson, 201536	UK	RCT	320 children, 4–11y, recent onset of ear symptoms and OME in one or both ears. Primary care	Nasal balloon autoinflation (160)	Usual care (160)		Moderate~	Adjusted between-group difference in change from baseline: −0.42 (−0.63 to −0.22) Conclusions: Compared with usual care, autoinflation produces greater improvements in ear-related quality of life in older children with recent onset of ear symptoms and OME.
Adhesive OM/ ET dysfunction
Si, 201837		RCT	120 patients, 15–75y, adhesive OM with conductive hearing loss. Tertiary care	ETBD + CT (30)	ETBD (30) CT (30) Conservative treatment; intranasal steroids and autoinflation using Valsalva (30)	COMOT-15 (0- 75)	High*	Mean postoperative COMOT-15 scores significantly decreased 3 mo, 6mo, 1 yr and 2 yrs after surgery in all three surgery groups and did not significantly change in the conservative treatment group. The post-operative QoL was significantly higher in the ETBD + CT than the CT only group. Conclusions: Cartilage tympanoplasty combined with ET balloon dilatation could improve QoL in patients with adhesive OM.

AOM: acute otitis media; COMOT-15: Chronic Otitis Media Outcome Test-15; CSOM: CT: cartilage tympanoplasty; d: days, ET: Eustachian tube; ETBD: Eustachian tube balloon dilatation; ENT: ear, nose, and throat; HUI3: The Health Utilities Index Mark 3; mo: months, OM: otitis media; OME: otitis media with effusion; PedsQL: The Pediatric Quality of Life Inventory; QoL: quality of life; RAOM: recurrent acute otitis media RCT: randomized controlled trial; VT: ventilation tubes; y: years;

¹Oral prednisolone: 20 mg/kg/d for children 2–5y or 30mg/kg/d for children 6–8y.

^{^}Risk of bias was downgraded from low to high due to study limitations (high risk of attrition bias and use of non-validated questionnaire)

^~Risk of bias was downgraded from low to moderate due to study limitations (unblinded nature of trial which may have impacted subjective outcome assessment)

^*Risk of bias was downgraded from low to high due to study limitations (high risk of bias due to sample size and unblinded nature of trial which may have impacted subjective outcome assessment)

We found two observational studies comparing the impact of two different treatment regimens on children’s QoL. One compared the use of narrow-versus broad-spectrum antibiotics for AOM without finding any differences in QoL between the groups [38]. The other compared two surgical procedures for chronic disease cholesteatoma that most often is a long-term development from especially chronic OME (Table 7) [33]. Also in this study, no differences in QoL were observed between groups.

Table 7.

Observational studies on impact of treatment for OM on QoL

Author, Year	Country	Study design	Participants, Setting	Intervention (participants)	Comparator (participants)	QoL Outcome (range)	Risk of Bias	Results and Conclusions
AOM
Gerber, 201738	USA	Cohort	1079 children, 6mo-12y, AOM Primary care	Narrow-spectrum antibiotics (411)	Broad-spectrum antibiotics (668)	PedsQL (0 to 100)	High’	Mean difference (full matched analysis) in PedsQL: -1.4 (-3.0 to 0.2) Conclusions: Broad-spectrum antibiotics were not associated with better QoL compared with narrow-spectrum antibiotics
CSOM/ Cholesteatomatous OM
Lucidi, 201933	Italy	Cohort	81 adults, 18–70y, cholesteatomatous OM Tertiary care	CWD mastoidectomy (50)	CWU mastoidectomy (31); limited extend of disease	COS (13 items, 3 subscales, 0–100) COMOT-15 (total scores were normalised to scale 0–100)	High’	CWD vs CWU COS: - 3 mo: 72 (SD 18) vs 74 (SD 14) - 12 mo: 79 (SD 14) vs 81 (SD 11) COMOT-15: - 12 mo: 80 (SD 19) vs 88 (SD 16) Conclusions: Postoperatively, no significant differences were observed between CWD and CWU in mean overall COS and COMOT-15 scores.

AOM: acute otitis media; COMOT-15: Chronic Otitis Media Outcome Test-15; COS: Chronic Ear Survey; ; CSOM: chronic suppurative otitis media; CWD: canal wall down; CWU: canal wall up; GBI: Glasgow Benefit Inventory; mo: months, PedsQL: The Pediatric Quality of Life Inventory; QoL: quality of life; y: years;

^‘Risk of bias was judged high due to observational study design

Implications for Clinical Practice and Future Research Goals

The quality of clinical research is highly dependent on study design, but also on the quality of the applied measurement instruments [39–41]. Hence, high quality clinical studies measuring subjective parameters such as QoL need to apply outcome measures in which relevant psychometric properties have been adequately established. In other words, assessment of these properties becomes crucial in assuring that the applied measurement instrument is capable of assessing what it purports to assess in a valid fashion. Therefore, assessment of internal consistency, validity and reproducibility is often required as minimum standards. Furthermore, if the measure is to be applied in longitudinal research assessment of its sensitivity to measuring change (responsiveness) is also necessary. Lastly, to enhance the applicability of outcome measures in clinical research, the user will want to know what a certain score means (interpretability). Table 8 shows an overview of relevant instrument properties for researchers planning a new study and as a tool for readers who wish to critically assess the methods used in the current literature.

Table 8.

Overview of relevant instrument properties for researchers planning a new study and as a tool for readers who wish to critically assess the methods used in the current literature

Internal consistency	The extent to which items in a scale or subscale are inter-correlated, thus measuring the same construct
Reliability/Reproducibility	Reproducibility concerns the degree to which repeated measurements in stable persons provide similar answers
Validity	The degree to which an instrument measures the construct(s) it purports to measure
Responsiveness (sensitivity to measuring change)	The ability of an instrument to detect change over time in the construct to be measured
Interpretability	The degree to which one can assign qualitative meaning to quantitative scores

Based on this review and general comments on studies of QoL and development concerning OM the following needs to be addressed in future studies:

1. It is highly recommended to routinely assess the impact on both QoL and hearing when evaluating interventions, whether medical, surgical or alternative remedies in children with OM.

2. Studies of QoL symptom score reversibility after treatment or other interventions are wanted along with temporal changes in QoL after OM.

3. The large variety of QoL questionnaires used (see Table 2) as well as the lack of an international reporting standard for hearing outcomes in studies of childhood OM add to the high complexity of trying to synthesize the available evidence on this topic.

4. There is an urgent need for more robust and uniform studies in this field and we would support the development of a core outcome set for the various OM entities that should include the most reliable and meaningful QoL and developmental outcome measurement instruments.

5. The literature still lacks high quality randomized studies. Future studies should include randomization while at the same time focus on including children from the whole spectrum of disease severity in order to maintain generalizability.