Abstract
Objective
To evaluate the cost-effectiveness of tympanostomy tube placement vs nonsurgical medical management, with the option of tympanostomy tube placement in the event of treatment failure, in children with recurrent acute otitis media (AOM).
Study design
A Markov decision model compared management strategies in children ages 6-35 months, using patient-level data from a recently completed, multicenter, randomized clinical trial of tympanostomy tube placement vs medical management. The model ran over a 2-year time horizon using a societal perspective. Probabilities, including risk of AOM symptoms, were derived from prospectively collected patient diaries. Costs and quality-of-life measures were derived from the literature. We performed one-way and probabilistic sensitivity analyses, and secondary analyses in predetermined low- and high-risk subgroups. The primary outcome was incremental cost per quality-adjusted life-year gained.
Results
Tympanostomy tubes cost $989 more per child than medical management. Children managed with tympanostomy tubes gained 0.69 more quality-adjusted life-days than children managed medically, corresponding to $520 855 per quality-adjusted life-year gained. Results were sensitive to the costs of oral antibiotics, missed work, special childcare, the societal cost of antibiotic resistance, and the quality of life associated with AOM. In probabilistic sensitivity analyses, medical management was favored in 66% of model iterations at a willingness-to-pay threshold of $100 000/quality-adjusted life-year. Medical management was preferred in secondary analyses of low- and high-risk subgroups.
Conclusions
For young children with recurrent AOM, the additional cost associated with tympanostomy tube placement outweighs the small improvement in quality of life. Medical management for these children is an economically reasonable strategy.
Trial registration
ClinicalTrials.gov number, NCT02567825.
Acute otitis media (AOM) is the most common pediatric infection for which antibiotics are prescribed in the US.1 Recurrent AOM, defined as 3 or more episodes of AOM in a 6 month period or 4 in a 12 month period, is associated with considerable economic burden and decreased quality of life.2–5 There was no difference in the yearly rate of episodes of AOM over a 2-year period among children with recurrent AOM who were randomized to surgical intervention with placement of tympanostomy tubes or to nonsurgical medical management with tympanostomy tube placement in the event of treatment failure.6 An aim of that study was to collect detailed data during the 2-year follow-up period that would allow a comparison of the costs and associated quality of life for each of the management strategies.
In this investigation, we use decision modeling techniques to evaluate the cost-effectiveness of tympanostomy tube placement and medical management of recurrent AOM in young children using data collected in the above-mentioned trial. In a secondary analysis, we investigated the cost-effectiveness of each strategy in high- or low-risk subgroups.
Methods
Study Design
We created a decision-analytic Markov model to simulate a hypothetical cohort of children 6-35 months of age with a history of recurrent AOM. We collected data prospectively as part of the randomized clinical trial, Efficacy of Tympanostomy-tubes for Children With Recurrent AOM (ClinicalTrials.gov number, NCT02567825).7 The protocol, which included the design for the cost-effectiveness analysis conducted here, was approved by the participating institutional review boards at University of Pittsburgh Medical Center Children’s Hospital of Pittsburgh; Children’s National Medical Center in Washington, D.C.; and Kentucky Pediatric and Adult Research in Bardstown, Kentucky. The decision model was programmed in TreeAge Pro 2016 (TreeAge Software, Inc).
We used intention-to-treat data for our primary analysis. Using these data, we constructed a hypothetical cohort of 1000 children and followed them over a 2-year period. We chose a 2-year time horizon because the trial followed enrolled children for 2 years and collected detailed data on costs and resource utilization. Per the trial protocol, parents used an electronic diary to record, on a daily basis, their child’s symptoms and any instance of adverse drug events. Use of medical and nonmedical resources was recorded during the trial.7
We considered the following health states: (1) well, (2) symptomatic with AOM, (3) symptomatic with otorrhea, or (4) hospitalized. Children in the simulated population were initially categorized as well and assigned to either tympanostomy tube placement or medical management with tympanostomy tubes placed in the event of treatment failure. Children who underwent tympanostomy tube placement did so at the start of the model. Over time, children in each strategy experienced episodes of AOM or otorrhea, underwent medical evaluations, received antibiotic treatment and additional medical procedures, and their caregivers missed work or required special childcare secondary to their children’s illnesses. Frequency of these events was drawn from trial data described above. To calculate days with AOM or otorrhea, we used patient diaries that were prospectively collected whenever patients were feeling unwell during the 2-year follow-up period. Days with AOM symptoms were defined by the presence of any symptom of AOM in a child with otoscopically confirmed AOM (ie, a score of ≥1 on the acute otitis media severity of symptom scale, version 4.0, for that day). Otorrhea and antibiotic use were also recorded daily on the electronic diary. Accordingly, quality-adjusted life days (QALDs) were calculated for each child included in the study using data provided prospectively by parents in daily diaries.
Model Inputs
Input measures for probabilities, costs, and outcomes are presented in Tables I and II. For variables based on clinical trial data and for variables drawn from the literature, we included a 95% probability range. For cost variables or variables for which a probability range was not reported, a probability range was estimated.
Table I.
Costs and utilities
| Variable | Point value | Range |
|---|---|---|
| Costs ($) | ||
| Hour of work8 | 29 | 15-45 |
| Primary care office visit9–12 | 157 | 36-307 |
| Urgent care visit9,13 | 270 | 250-511 |
| Otolaryngology office visit11,12 | 279 | 200-400 |
| Emergency department visit9,14 | 520 | 385-555 |
| Tympanostomy-tube placement procedure12,15 | 739 | 700-2065 |
| Adenectomy16,17 | 1151 | 480-1822 |
| Tonsillectomy16–18 | 1222 | 480-2385 |
| Adenotonsillectomy16,17 | 1355 | 1029-2385 |
| Ear wick19 | 7 | 5-9 |
| Hospitalization9,20 | 1630 | 1205-13 236 |
| Ibuprofen oral suspension, per mL21 | 0.04 | 0.03-0.08 |
| Nystatin topical cream, per gm21 | 0.42 | 0.35-0.80 |
| Ofloxacin 0.3% otic suspension, per mL21 | 17 | 14-20 |
| Amoxicillin-clavulanate oral suspension, per mL21 | 0.20 | 0.06-0.51 |
| Societal cost of antibiotic resistance, per prescription22 | 14 | 3-95 |
| Utilities | ||
| Acute otitis media10 | 0.79 | 0-100 |
| Otorrhea | 0.90 | 0-100 |
| Diarrhea10 | 0.88 | 0-100 |
| Diaper dermatitis10 | 0.99 | |
| Hospitalization23 | 0.88 | 0-100 |
| Other | ||
| Hours in a workday | 8 | 6-12 |
| Cycle length | 1 day |
Table II.
Clinical parameters used in the base case model
| Variable | Medical management |
Tympanostomy-tubes |
||
|---|---|---|---|---|
| Point value | 95% CI | Point value | 95% CI | |
| Proportion of children undergoing tympanostomy-tube placement | 0.45 | 0.36, 0.53 | 0.90 | 0.85, 0.95 |
| Proportion of children undergoing second tympanostomy-tube placement | 0.02 | 0.00, 0.05 | 0.09 | 0.04, 0.15 |
| Days per child per y | ||||
| Otorrhea | 2.83 | 1.28, 4.37 | 7.96 | 5.79, 10.14 |
| AOM-SOS ≥1 | 8.33 | 7.17, 9.49 | 2.00 | 1.44, 2.57 |
| Protocol-defined diarrhea* | 0.79 | 0.48, 1.10 | 0.60 | 0.22, 0.97 |
| Diaper dermatitis† | 1.55 | 0.94, 2.16 | 0.99 | 0.48, 1.49 |
| Systemic antibiotics | 12.92 | 11.13, 14.71 | 8.76 | 6.89, 10.63 |
| Otic antibiotic drops | 6.92 | 4.37, 9.48 | 19.43 | 15.61, 23.26 |
| Oral analgesics | 1.27 | 0.83, 1.71 | 0.90 | 0.44, 1.35 |
| Antifungal cream | 1.37 | 0.88, 1.85 | 0.96 | 0.46, 1.47 |
| Hospitalization | 0.01 | 0.00, 0.03 | 0.00 | 0.00, 0.01 |
| Outpatient medical encounters per child per y | ||||
| Primary care physician | 2.13 | 1.92, 2.35 | 2.06 | 1.83, 2.28 |
| Urgent care | 0.34 | 0.24, 0.44 | 0.36 | 0.27, 0.44 |
| Emergency department | 0.44 | 0.32, 0.57 | 0.54 | 0.39, 0.69 |
| Otolaryngology (excluding postoperative follow-up) | 0.24 | 0.16, 0.32 | 0.34 | 0.24, 0.43 |
| Days missed work for illness per child per y | 2.73 | 2.04, 3.42 | 2.71 | 2.18, 3.24 |
| Days special childcare for illness per child per y | 2.56 | 1.84, 3.27 | 3.16 | 2.38, 3.94 |
| Additional otolaryngology procedures per study period | ||||
| Adenoidectomy | 1 | 0, 2 | 7 | 5, 9 |
| Tonsillectomy | 0 | 0, 1 | 1 | 0, 2 |
| Adenotonsillectomy | 1 | 0, 2 | 1 | 0, 2 |
| Wick placement | 2 | 1, 3 | 3 | 2, 4 |
AOM-SOS, acute otitis media severity of symptom scale.
Protocol-defined diarrhea was the occurrence of at least 3 watery stools on 1 day or at least 2 watery stools on 2 consecutive days.
Diaper dermatitis was defined as dermatitis resulting in topical antifungal treatment.
Costs included direct costs of medical evaluations, diagnostic testing, and medical treatment. We based costs of outpatient medical evaluations and procedures on prior publications,9–19 and costs of hospitalization on prior publications and median value for pediatric hospitalizations, as reported by the Health care Cost and Utilization Project.9,20 We based medication costs on the National Average Drug Acquisition Cost, as reported by the Centers for Medicare & Medicaid Services.21 Each prescription for systemic antibiotics was assigned an additional cost to account for the societal cost of antibiotic resistance associated with antibiotic use in the US22 Costs of missed work and special childcare were based on average hourly wages from the US Bureau of Labor Statistics.8 We adjusted costs to 2020 US dollars based on the medical cost component of the Consumer Price Index,24 and selected 2020 as this was the last year of clinical trial data collection.
Health state utility values were drawn from the literature and QALDs were calculated as the duration in health states multiplied by the utility of that state.10,23 We used an estimate of quality-of-life utility for otorrhea of 0.9, as no data were available in the literature. Costs and utilities were not discounted given the 2-year time horizon.
Cost-effectiveness Analysis
We conducted our cost-effectiveness analysis from a societal perspective, considering health expenditure costs as well as missed work and additional child care.25 Primary outcomes we evaluated included cost, effectiveness (days without symptoms of AOM), and cost-effectiveness. We ranked strategies by cost then compared them in terms of cost, effectiveness, and incremental cost-effectiveness ratio (ICER). We assumed a willingness-to-pay of $100 000/quality-adjusted life-year (QALY) gained a commonly cited threshold for the US health care system.25
Subgroup Analysis
To determine whether the cost-effectiveness of tympanostomy tubes may be different for children at higher risk of recurrent AOM, we conducted a secondary analysis focusing on high- and low-risk cohorts. We determined risk level a priori using a 16-point scale that assessed age of onset of AOM and quantity and severity of prior infections. We defined high-risk for recurrent AOM as a score ≥8. This scale is described in detail in the final version of the protocol.6
Sensitivity Analyses
We conducted one-way sensitivity analyses to determine if varying any individual measure across its listed range substantially changed results. Threshold analyses were performed to determine the point at which changes to certain input measures resulted in a substantial change in the preferred strategy. We then performed probabilistic sensitivity analyses in which all measures could vary simultaneously; we ran these models 1000 times and report findings as a cost-effectiveness acceptability curve,26 which shows the likelihood that strategies would be favored over a range of cost-effectiveness willingness-to-pay (or acceptability) thresholds from $0 to $200 000/QALY gained. For the probabilistic sensitivity analysis, we randomly chose values for each measure from the provided distributions. We used β distributions for probabilities and quality adjustments and γ distributions for costs.
Results
In the primary analysis, medical management was the less expensive strategy (cost approximately $1000 less than tympanostomy tube placement over the 2-year period). The tympanostomytube placement strategy was more effective; children undergoing surgery had a gain of approximately 0.69 days of perfect health over 2 years. Accordingly, compared with the medical management strategy, the tympanostomy tube placement strategy cost an additional $1427/QALD gained (Table III and Table IV; available at www.jpeds.com). This approximates to $520 855 per QALY, which exceeds the commonly cited $100 000/QALY willingness-to-pay threshold for the US health care system.
Table III.
Cost-effectiveness analysis
| Strategy | Cost ($) | Incremental cost ($) | Effectiveness (QALDs) | Incremental effectiveness (QALDs gained) | ICER ($/QALD) | ICER ($/QALY) |
|---|---|---|---|---|---|---|
| Primary analysis | ||||||
| Medical management | 4497 | - | 725.07 | - | - | - |
| Tympanostomy-tubes | 5486 | 989 | 725.76 | 0.69 | 1427 | 520 855 |
| High risk cohort | ||||||
| Medical management | 4455 | - | 725.21 | - | - | - |
| Tympanostomy-tubes | 6008 | 1553 | 725.38 | 0.17 | 9065 | 3 308 725 |
| Low risk cohort | ||||||
| Medical management | 4433 | - | 724.99 | - | - | - |
| Tympanostomy-tubes | 4782 | 349 | 726.20 | 1.21 | 287 | 104 755 |
Table IV.
Breakdown of costs and utilities per person by strategy over 2 years, base-case analysis
| Variable | Medical management | Tympanostomy-tubes |
|---|---|---|
| Costs per person over 2 y ($) | ||
| Tympanostomy tube placement | 342 | 729 |
| Otic antibiotic drops | 117 | 330 |
| Systemic antibiotics | 41 | 28 |
| Societal cost of antibiotic resistance | 37 | 25 |
| Primary care office visits | 668 | 668 |
| Emergency department visits | 457 | 509 |
| Urgent care visits | 183 | 176 |
| Otolaryngology office visits | 134 | 196 |
| Hospitalizations | 40 | 13 |
| Special childcare | 1188 | 1466 |
| Missed work | 1266 | 1258 |
| Additional otolaryngologic surgery | 20 | 88 |
| Ear wick | <1 | <1 |
| Ibuprofen | 1 | 1 |
| Nystatin topical cream | 2 | 2 |
| TOTAL | 4497 | 5486 |
| Additional cost | 989 | |
| QALDs lost over 2 y | ||
| AOM symptoms | 4.14 | 2.47 |
| Otorrhea | 0.56 | 1.59 |
| Protocol-defined diarrhea | 0.19 | 0.16 |
| Diaper dermatitis | 0.03 | 0.02 |
| Hospitalization | 0.006 | 0.002 |
| TOTAL lost | 4.93 | 4.24 |
| QALDs experienced over 2 y* | ||
| Total QALD | 725.07 | 725.76 |
| Additional QALDs | 0.69 | |
| ICER (Cost/QALD) | $1427 | |
| ICER (Cost/QALY) | $520 855 |
Bold text indicates the additional cost or QALDs accrued with the tympanostomy-tubes strategy compared to medical management.
Calculated by subtracting total QALDs lost over the 2 years from the maximum possible QALD (ie, 730).
One-way sensitivity analyses demonstrated that the model was sensitive to the cost of oral antibiotics, the societal cost of antibiotic resistance, the quality of life associated with AOM, and the duration of missed work and special childcare associated with each strategy (Table V). The results of the probabilistic sensitivity analysis were consistent with the findings from the primary analysis; medical management was the preferred strategy at all willingness to pay thresholds from $0 to 200 000/QALY. At a willingness-to-pay threshold of $100 000/QALY, the medical management strategy was preferred in 66% of model iterations (Figure).
Table V.
One-way sensitivity analysis
| Variable | Base-case value | Threshold value | Preferred strategy |
|
|---|---|---|---|---|
| Below threshold | Above threshold | |||
| Cost of oral antibiotics, per d | 1.60 | 97.90 | Medical management | Tympanostomy-tubes |
| Societal cost of antibiotic resistance, per prescription | 14.45 | 97.74 | Medical management | Tympanostomy-tubes |
| Utility of acute otitis media | 0.79 | 0.42 | Tympanostomy-tubes | Medical management |
| D of special child care per y | ||||
| Medical management | 2.56 | 4.28 | Medical management | Tympanostomy-tubes |
| Tympanostomy-tubes | 3.16 | 1.44 | Tympanostomy-tubes | Medical management |
| D of missed work per y | ||||
| Medical management | 2.73 | 4.45 | Medical management | Tympanostomy-tubes |
| Tympanostomy-tubes | 2.71 | 0.99 | Tympanostomy-tubes | Medical management |
Figure.
Probabilistic sensitivity analysis. Results are shown as cost-effectiveness acceptability curves, depicting cumulative probability (y-axis) that the cost-effectiveness of a strategy will be less than or equal to a given willingness-to-pay threshold (x-axis).
Subgroup Analyses
In the subgroup analysis, medical management was the preferred strategy for individuals at high-risk of recurrent AOM. Tympanostomy tube placement cost an additional $1553 for an additional 0.17 QALDs per patient over 2 years, resulting in an ICER of $9065 per QALD gained. The ICER in terms of cost per QALY gained exceeded the $100 000 per QALY gained willingness-to-pay threshold (Table III). One-way sensitivity analyses demonstrated that this model was sensitive to parental missed work and special childcare required (Table VI; available at www.jpeds.com). Probabilistic sensitivity analyses demonstrated that at a willingness-to-pay threshold of $100 000/QALY, medical management was the preferred strategy for patients in the high-risk cohort in 76% of model iterations.
Table VI.
Subgroup one-way sensitivity analyses
| Variable | Base-case value | Threshold value | Preferred strategy |
|
|---|---|---|---|---|
| Below threshold | Above threshold | |||
| High-risk cohort | ||||
| D of missed work per y | ||||
| Medical management | 2.84 | 6.20 | Medical management | Tympanostomy-tubes |
| D of special childcare per y | ||||
| Medical management | 2.62 | 6.09 | Medical management | Tympanostomy-tubes |
| Tympanostomy-tubes | 3.59 | 0.01 | Tympanostomy-tubes | Medical management |
| Low-risk cohort | ||||
| Cost of oral antibiotics, per d | 1.60 | 2.98 | Medical management | Tympanostomy-tubes |
| Societal cost of antibiotic resistance, per prescription | 1.45 | 2.82 | ||
| Cost of otic antibiotics per d | 8.5 | 7.7 | Tympanostomy-tubes | Medical management |
| Cost of tympanostomy procedure | 739 | 706 | Tympanostomy-tubes | Medical management |
| Cost of emergency department visit | 520 | 452 | ||
| Cost of urgent care visit | 270 | 373 | ||
| Utility of acute otitis media | 0.79 | 0.78 | Tympanostomy-tubes | Medical management |
| Utility of otorrhea | 0.90 | 0.91 | Medical management | Tympanostomy-tubes |
| Utility of diaper dermatitis | 0.99 | 0.96 | ||
| Utility of diarrhea | 0.88 | 0.81 | ||
| D of missed work per y | ||||
| Medical management | 2.66 | 2.70 | Medical management | Tympanostomy-tubes |
| Tympanostomy-tubes | 2.09 | 2.05 | Tympanostomy-tubes | Medical management |
| D of special childcare per y | ||||
| Medical management | 2.52 | 2.56 | Medical management | Tympanostomy-tubes |
| Tympanostomy-tubes | 2.53 | 2.49 | Tympanostomy-tubes | Medical management |
In the low-risk cohort, tympanostomy tube placement cost an additional $349 per patient with 1.2 additional QALDs over 2 years, yielding an ICER of $287/QALD gained. One-way sensitivity analyses demonstrated that this model was sensitive to the costs of oral and otic antibiotics, tympanostomy tube placement, missed work, and special childcare; and the quality of life associated with both AOM and otorrhea (Table VI). The ICER exceeded the $100 000 per QALY gained willingness-to-pay threshold. For the low-risk cohort, medical management was the preferred strategy in 51% of model iterations.
Discussion
We performed a cost-effectiveness analysis to compare tympanostomy tube placement and medical management in children with recurrent AOM and found that medical management provided economically reasonable outcomes compared with tympanostomy tube placement. Medical management was the preferred strategy in both high- and low-risk subgroups. The main determinants of these results were the higher cost for tympanostomy tube placement, childcare, and otic antimicrobial agents in the tympanostomy tube group which were not accompanied by a similar increase in each child’s quality of life.
We found that model results were robust. Although results were sensitive to the cost of oral antibiotics, the utility of AOM, and the duration of missed work and special childcare, in every case, the threshold values needed for surgical management to become the favored strategy was far from the base case values, which, in almost all cases, were derived directly from the trial data or from the literature.
We included the societal cost of antimicrobial resistance in this model. Studies examining community antimicrobial resistance suggest that antibiotic prescriptions may contribute to the development of antibiotic resistant pathogens.27 The true cost of resistance on an individual prescription basis is difficult to assess. Our selection of $14 per prescription was based on the literature and assigned on a per prescription basis.22 There were more antibiotic prescriptions in the medical management strategy. Thus, the model included a higher burden of societal cost of antibiotic resistance for this strategy. In the trial from which data for this study were drawn, there were no significant between-group differences in the percentage of children with nasopharyngeal colonization with any penicillin-nonsusceptible pathogen. Moreover, conditional odds ratios, measuring community-wide effects antibiotic treatment on antibiotic resistance, did not differ between the treatment groups, suggesting that societal impact of antibiotic prescriptions was not significantly different between the 2 treatment arms.28 In our model, one-way sensitivity analyses demonstrated that the societal cost of antibiotic resistance would need to be greater than $97 per prescription in order for the tympanostomy tubes management strategy to be preferred. Given the trial findings of no between-group differences in measures of the effect of antibiotic prescriptions on antimicrobial resistance, it is likely that the assignment of societal cost of antibiotic resistance by antibiotic prescription biased the model against the medical management strategy. Topical antibiotics also contribute to community antibiotic resistance, although likely on a smaller scale than systemic antibiotics.29 The decision to not assign a societal cost of antimicrobial resistance to these prescriptions again biased our results against the medical-management strategy. Despite this, the medical management strategy was the preferred strategy in all 3 versions of the model.
We found that patients in the predefined low-risk group saw the most benefit and lowest relative costs from tympanostomy tube placement, a finding which was counterintuitive. This outcome was impacted by differences in both cost and quality of life. Children assigned to the tympanostomy tubes strategy in the high-risk cohort required more medical and nonmedical resources than those assigned to the medical management strategy, and gained 0.17 QALDs, the equivalent of 4 symptom-free hours, in 2 years. In contrast, although those assigned to the tympanostomy tubes strategy in the low-risk cohort, did incur more costs of both medical and nonmedical resources than those assigned to medical management, the difference was smaller than that in the high-risk cohort. Low-risk patients in the tympanostomy tubes strategy also accrued 1.2 additional QALDs over the 2-year time period. However, the cost for this additional benefit exceeded the $100 000/QALY willingness-to-pay threshold.
By simulating the experience of large patient populations, Markov analyses identify strategies that benefit the most individuals most often and lend themselves to broader interpretations. Our study findings may also inform shared decision-making discussions. Previous studies have cited that illness symptoms, missed work, and use of medical resources contribute to the individual burden associated with recurrent AOM.5,30 Patients in both treatment strategies experienced subsequent episodes of AOM and use of both non-medical and medical resources. Individual experience of AOM symptoms or otorrhea may differ from that of a population, making either strategy more or less valuable on an individual basis.
This study has a number of strengths. Data were drawn from a randomized-control trial of young children with recurrent AOM, all of whom had received the pneumococcal conjugate vaccine, and included cost-effectiveness analysis as an aim. Episodes of AOM were diagnosed by trained otoscopists and treated according to a standardized protocol. Quality of life information was collected with daily electronic symptom diaries using validated scales for rating symptoms severity and functional outcomes. Resource utilization was documented at regular intervals, allowing for a comprehensive assessment of individual quality-of-life experiences.
This study has several limitations. We could not find a quality-of-life utility value for otorrhea in the literature. However, utility of otorrhea did not influence the conclusions of the primary analysis; at no utility value for otorrhea did tympanostomy tubes placement become the preferred strategy. Patients who did not complete the full 2 years of the original trial were assumed to have medical costs and quality of life that did not vary significantly from those experienced during trial participation, an assumption that could bias results toward either management strategy. The 2-year time horizon was based on the 2 years of trial data collected. Given the decreasing frequency of AOM with age and the self-limited nature of tympanostomy tubes, it was not reasonable to extrapolate beyond the 2 years available follow-up data. Additionally, the model does not account for long-term impacts of recurrent AOM in childhood. Given the age of patients enrolled, it is likely that the model captures the time period when individuals are most likely to experience AOM, and AOM to benefit from medical or surgical interventions. Adverse antibiotic events can include diarrhea, vomiting, candidiasis, non-candidal rash, other allergic reaction, and unspecified adverse events.31 Given the trial design, it is possible that adverse events other than protocol-defined diarrhea and diaper dermatitis which were not brought to medical attention may not have been included in the model. We attempted to include utilization of nonmedical resources by including both missed work and special childcare needs but were unable to account for the costs of informal childcare arrangements or missed school. Diagnosis of AOM in the trial was made by trained otoscopists. This model does not account for overdiagnosis of AOM, which is estimated to occur in up to 25% of AOM diagnoses.32 The trial was conducted in 2 large children’s hospitals and thus data may not be directly applicable to other settings.
The additional costs associated with the tympanostomy tube placement strategy and limited benefit suggest that non-surgical medical management is the economically preferred strategy.
Glossary
- AOM
Acute otitis media
- ICER
Incremental cost-effectiveness ratio
- QALD
Quality-adjusted life-day
- QALY
Quality-adjusted life-year
- US
United States
Footnotes
A.H. reports owning stock in Kaizen Bioscience, holding patent 9, 636, 007 B2 on a method and apparatus for aiding diagnosis of otitis media by classifying tympanic-membrane images, and holding patent 9, 987, 257 B2 on pediatric oral-suspension formulation of amoxicillin–clavulanate potassium and the method for use, licensed to Kaizen Bioscience. The other authors declare no conflicts of interest.
References
- 1.Finkelstein JA, Raebel MA, Nordin JD, Lakoma M, Young JG. Trends in outpatient antibiotic use in 3 health plans. Pediatrics 2019;143:e20181259. 10.1542/peds.2018-1259 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Lieberthal AS, Carroll AE, Chonmaitree T, Ganiats TG, Hoberman A, Jackson MA, et al. The diagnosis and management of acute otitis media. Pediatrics 2013;131:e964–99. 10.1542/peds.2012-3488 [DOI] [PubMed] [Google Scholar]
- 3.Oh PI, Maerov P, Pritchard D, Knowles SR, Einarson TR, Shear NH. A cost-utility analysis of second-line antibiotics in the treatment of acute otitis media in children. Clin Ther 1996;18:160–82. 10.1016/S0149-2918(96)80188-3 [DOI] [PubMed] [Google Scholar]
- 4.Rovers MM. The burden of otitis media. Vaccine 2008;26:10–2. 10.1016/j.vaccine.2008.11.005 [DOI] [PubMed] [Google Scholar]
- 5.Mui S, Rasgon BM, Hilsinger RL, Lewis B, Lactao G. Tympanostomy tubes for otitis media: quality-of-life improvement for children and parents. Ear, Nose Throat J 2005;84:418–24. 10.1177/014556130508400712 [DOI] [PubMed] [Google Scholar]
- 6.Hoberman A, Preciado D, Paradise JL, Chi DH, Haralam M, Block SL, et al. Tympanostomy tubes or medical management for recurrent acute otitis media. N Engl J Med 2021;384:1789–99. 10.1056/nejmoa2027278 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hoberman A, Preciado D, Paradise JL, Chi DH, Haralam M, Block SL, et al. Efficacy of tympanostomy tubes for children with recurrent acute otitis media. Identifier NCT02567825. Accessed July 11, 2022. https://clinicaltrials.gov/ct2/show/NCT02567825
- 8.Bureau of Labor Statistics, U.S. Department of Labor. The employment situation [news release]. 2021. Accessed August 15, 2022. https://www.bls.gov/news.release/archives/empsit_01082021.pdf
- 9.Tong S, Amand C, Kieffer A, Kyaw MH. Trends in healthcare utilization and costs associated with pneumonia in the United States during 2008-2014. BMC Health Serv Res 2018;18:1–10. 10.1186/s12913-018-3529-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Shaikh N, Dando EE, Dunleavy ML, Curran DL, Martin JM, Hoberman A, et al. A cost-utility analysis of 5 strategies for the management of acute otitis media in children. J Pediatr 2017;189:54–60.e3. 10.1016/j.jpeds.2017.05.047 [DOI] [PubMed] [Google Scholar]
- 11.Casey JR, Pichichero ME. Payment analysis of two diagnosis and management approaches of acute otitis media. Clin Pediatr 2014;53:865–73. 10.1177/0009922814533592 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Montalbano A, Rodean J, Kangas J, Lee B, Hall M. Urgent care and emergency department visits in the pediatric medicaid population. Pediatrics 2016;137:e20153100. 10.1542/peds.2015-3100 [DOI] [PubMed] [Google Scholar]
- 13.Hart RJ, Stevenson MD, Smith MJ, Scott LJA, Cross K. Cost-effectiveness of strategies for offering influenza vaccine in the pediatric emergency department. JAMA Pediatr 2018;172:e173879. 10.1001/jamapediatrics.2017.3879 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Raol N, Zogg CK, Boss EF, Weissman JS. Inpatient pediatric tonsillectomy: does hospital type affect cost and outcomes of care? Otolaryngol–Head Neck Surg 2016;154:486–93. 10.1177/0194599815621739 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Sjogren PP, Thomas AJ, Hunter BN, Butterfield J, Gale C, Meier JD. Comparison of pediatric adenoidectomy techniques. Laryngoscope 2018;128:745–9. 10.1002/lary.26904 [DOI] [PubMed] [Google Scholar]
- 16.Sjogren PP, Gale C, Henrichsen J, Olsen G, Ott MJ, Peters M, et al. Variation in costs among surgeons and hospitals in Pediatric tympanostomy tube placement. Laryngoscope 2016;126:1935–9. 10.1002/lary.25775 [DOI] [PubMed] [Google Scholar]
- 17.Pond F, McCarty D, O’Leary S. Randomized trial on the treatment of oedematous acute otitis externa using ear wicks or ribbon gauze: clinical outcome and cost. J Laryngol Otol 2002;116:415–9. 10.1258/0022215021911130 [DOI] [PubMed] [Google Scholar]
- 18.Meier JD, Zhang Y, Greene TH, Curtis JL, Srivastava R. Variation in pediatric outpatient adenotonsillectomy costs in a multihospital network. Laryngoscope 2015;125:1215–20. 10.1002/lary.24981 [DOI] [PubMed] [Google Scholar]
- 19.Davis KE, Carper K. Statistical Brief #381: Use and expenses for office-based physician visits by specialty, 2009: estimates for the U.S. civilian noninstitutionalized population, 2012, Surv Agency Healthc Res Qual Med Expend Panel. Accessed February 3, 2022. https://www.meps.ahrq.gov/data_files/publications/st381/stat381.shtml [Google Scholar]
- 20.Healthcare cost and Utilization Project, Agency for healthcare research and quality, September 2021. 2019. Accessed February 3, 2022. https://hcupnet.ahrq.gov [Google Scholar]
- 21.NADAC (national average druge acquisition cost), Centers for Medicare & Medicaid Services, August 2021. 2021. Accessed February 3, 2022. https://data.medicaid.gov/Drug-Pricing-and-Payment/NADAC-National-Average-Drug-Acquisition-Cost-/a4y5-998d [Google Scholar]
- 22.Michaelidis CI, Fine MJ, Lin CJ, Linder JA, Nowalk MP, Shields RK, et al. The hidden societal cost of antibiotic resistance per antibiotic prescribed in the United States: an exploratory analysis. BMC Infect Dis 2016;16:1–8. 10.1186/s12879-016-1990-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Hoshi SL, Kondo M, Okubo I. Economic evaluation of immunisation programme of 23-valent pneumococcal polysaccharide vaccine and the inclusion of 13-valent pneumococcal conjugate vaccine in the list for single-dose subsidy to the elderly in Japan. PLoS One 2015;10:1–16. 10.1371/journal.pone.0139140 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.CPI Inflation Calculator. CPI inflat calc 2019:1–2. Accessed July 9, 2022. https://data.bls.gov/cgi-bin/cpicalc.pl
- 25.Sanders G, Neumann P, Basu A, Brock D, Feeny D, Krahn M, et al. Recommendations for conduct, methodological practices, and reporting of cost-effectiveness analyses: second panel on cost-effectiveness in health and medicine. J Am Med Assoc 2016;316:1093–103. 10.1001/jama.2016.12195 [DOI] [PubMed] [Google Scholar]
- 26.Stinnett AA, Mullahy J. Net health benefits: a new framework for the analysis of uncertainty in cost-effectiveness analysis. Med Decis Mak 1998;18:68–80. [DOI] [PubMed] [Google Scholar]
- 27.Patrick DM, Hutchinson J. Antibiotic use and population ecology: how you can reduce your “resistance footprint”. CMAJ 2009;180:416–21. 10.1503/cmaj.080626 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Lipsitch M Measuring and interpreting associations between antibiotic use and penicillin resistance in Streptococcus pneumoniae. Clin Infect Dis 2001;32:1044–54. 10.1086/319604 [DOI] [PubMed] [Google Scholar]
- 29.Noonan KY, Kim SY, Wong LY, Martin IW, Schwartzman JD, Saunders JE. Treatment of ciprofloxacin-resistant ear infections. Otol Neurotol 2018;39:E837–42. 10.1097/MAO.0000000000001966 [DOI] [PubMed] [Google Scholar]
- 30.Rosenfeld RM, Tunkel DE, Schwartz SR, Anne S, Bishop CE, Chelius DC, et al. Clinical practice guideline: tympanostomy tubes in children (update). Otolaryngol–Head Neck Surg (United States) 2022;166:S1–55. 10.1177/01945998211065662 [DOI] [PubMed] [Google Scholar]
- 31.Gerber JS, Ross RK, Bryan M, Localio AR, Szymczak JE, Wasserman R, et al. Association of broad- vs narrow-spectrum antibiotics with treatment failure, adverse events, and quality of life in children with acute respiratory tract infections. J Am Med Assoc 2017;318:2325–36. 10.1001/jama.2017.18715 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Legros JM, Hitoto H, Garnier F, Dagorne C, Parot-Schinkel E, Fanello S. Clinical qualitative evaluation of the diagnosis of acute otitis media in general practice. Int J Pediatr Otorhinolaryngol 2008;72:23–30. 10.1016/j.ijporl.2007.09.010 [DOI] [PubMed] [Google Scholar]