Trends in Tympanostomy Tube Insertion Among Children Aged 0–12 Years: A Register-Based Cohort Study in Denmark (2003–2023)

Abstract

Purpose

Tympanostomy tube insertion (TTI) is one of the most frequently performed surgical procedures in children. Yet, substantial temporal and geographical variations persist across comparable healthcare systems. We aimed to describe nationwide trends in TTI among Danish children in 2003–2023, with particular attention to changes associated with the COVID-19 pandemic and regional differences in treatment patterns.

Patients and methods

In this nationwide register-based cohort study, we included all Danish children aged 0–12 years between 2003 and 2023 (n=2,243,285). Data on TTIs were obtained from the Danish National Patient Register and the Danish National Health Service Register. Incidence rates, cumulative proportions, age at first procedure, and regional variation were analyzed using calendar year and birth cohorts.

Results

We identified 704,013 TTI procedures performed in 371,818 children; 96.6% were conducted in ear, nose, and throat (ENT) practices. Annual procedure numbers peaked at 40,826 cases in 2010 and declined steadily to 29,846 cases in 2019. A marked reduction occurred during the COVID-19 pandemic in 2020–2021, followed by a rebound to the pre-pandemic level in 2022 and a further increase to 35,024 procedures in 2023. The proportion of children receiving at least one TTI increased from 3.11% in 2019 to 3.67% in 2023 (proportion ratio (PR) 1.18, 95% confidence interval (CI) 1.16–1.20). Median age at first TTI was 18.1 months, with persistent regional variation; rates were highest in the North Denmark Region and lowest in the Capital Region of Denmark.

Conclusion

TTI rates in Danish children decreased substantially over the past decade, declined further during the COVID-19 pandemic, and subsequently returned to pre-pandemic levels. Persistent regional differences indicate variation in clinical practice that cannot be explained solely by underlying disease epidemiology. The long-term effects of the pandemic on TTI utilization remain uncertain.

Plain Language Summary

Why was this study done?

This study was done to understand how often tympanostomy tube insertion (TTI), a common ear surgery procedure for middle ear infections in children, is performed in Denmark. We wanted to examine changes in surgery rates over time, and how the COVID-19 pandemic affected these procedures. We also wanted to investigate whether regional differences might reflect variations in medical practice.

What was found?

TTI rates in Danish children peaked around 2010 and then gradually declined until 2019. During the COVID-19 pandemic, the number of surgeries dropped sharply. The number returned to pre-pandemic levels by 2022 and rose further in 2023. The typical age at first TTI was 18 months. Surgery rates differed by region; the highest rates were seen in the North Denmark Region and the lowest in the Capital Region of Denmark.

What do these results mean?

TTI surgery has declined over the past decade, specifically during the pandemic, but it is now rising again. Regional differences suggest that local medical practices may influence whether children receive TTI. The long-term impact of the pandemic on these procedures remains uncertain.

Introduction

Otitis media (OM) is one of the most frequently diagnosed conditions in childhood and a leading cause of healthcare utilization in Western countries,^1,2 including Denmark.³ OM presents as a spectrum of symptoms, ranging from OM with effusion (OME) to severe acute otitis media (AOM) with complications. Otitis media with effusion (OME) is an accumulation of fluid in the middle ear cavity, often occurring secondary to an upper respiratory tract infection. The associated hearing loss is temporary and persists only while the effusion is present. Nevertheless, this may lead to delayed speech development in children, social difficulties, poor sleep, and reduced quality of life.^3–5 OM often causes recurrent sick days, affecting not only the child’s health but also parental work attendance and well-being.^5,6 In contrast, AOM can present with symptoms ranging from mild earache and low-grade fever to more severe clinical manifestations, dominated by otalgia, high fever, and gastrointestinal symptoms.^2,7 Complications to AOM, such as meningitis, mastoiditis, or sepsis, are rare but serious.⁸

Management of OME includes watchful waiting and autoinflation (ie, Valsalva’s maneuver; Otovent may be used in older children who are able to cooperate), whereas AOM is typically treated with analgesics and, when indicated, antibiotics or myringotomy.⁷ Recurrent AOM (rAOM) and chronic OM with effusion (cOME) may, following appropriate observation time in cases without spontaneous resolution, be managed with tympanostomy tube insertion (TTI).^7,9

Consequently, TTI is among the most frequently performed surgical procedures in children in Western countries, including Denmark.^10,11 In 2012, approximately 35,000 TTIs were performed in Denmark. This corresponds to an incidence rate of 40 per 1000 children.¹¹ Countries with comparable populations and similar healthcare systems have reported substantially lower TTI incidence, eg, 7 per 1000 in Sweden, 12 per 1000 in Norway, and 20 per 1000 in the Netherlands.^12,13 Thus, considerable variations in the incidence of TTI procedures are well documented, even when the epidemiology of OM is expected to be comparable.^12,14

During the COVID-19 pandemic, the incidence of AOM and OME declined markedly, leading to a significant pandemic-related reduction in TTI.¹⁵ In a Finnish national register spanning 2012–2022, the number of TTI procedures fell to 44.5% in 2020 and to 67.6% in 2021 relative to the pre-pandemic baseline level, returning to expected levels by 2022; this suggests both a strong effect of mitigation measures and a subsequent rebound as restrictions eased.¹⁶ Multi-center pediatric data from the United States are consistent with these observations, demonstrating sharp declines in OM encounters and myringotomy/TTI procedures from the spring of 2020.¹⁷ However, changes in TTI rates during and after the COVID-19 pandemic have not been investigated in Denmark.

This study aims to describe trends in TTI among Danish children from 2003 to 2023, with a particular focus on changes associated with the COVID-19 pandemic. Using national registry data, we quantified procedure rates over time, stratified by age group, geographical region, and healthcare sector, to assess whether the pandemic caused a transient disruption or a sustained shift in procedure rates. Additionally, we compared post-pandemic procedure rates with pre-pandemic baseline rates to assess the extent of rebound or persistent reduction in surgical activity.

Methods

Data Sources

The Danish healthcare system is publicly funded through taxation, ensuring equal access to care for all residents.^18,19 Danish registers provide complete and validated high-quality data on age, sex, vital status, and geographical region.^20,21 This structure provides a foundation for capturing data on TTI procedures through the two national health registers described below.

The Danish National Patient Register (DNPR) was established in 1977. Since then, all admissions to public hospitals and discharge diagnoses have been recorded here. The inclusion of data from private hospitals was fully implemented in 2003. The DNPR is recognized as one of the most comprehensive and reliable health registers globally.²² For otologic surgical procedures, the positive predictive value of coding, defined as the likelihood that a coded procedure represents the actual surgical intervention performed, has been reported to be 89%.²⁰

In Denmark, patients may consult otorhinolaryngologists in ear, nose, and throat practices (hereafter referred to as ENT practices) without referral from a general practitioner or hospital. These ENT practices perform minor surgical ambulatory procedures, such as TTI and adenoidectomy, under general anesthesia. The Danish National Health Service Register (NHSR) has collected data from primary healthcare, including ENT practices, since 1990. As all residents in Denmark are covered by the national, free-of-charge health insurance system, these data are generated through reimbursement claims submitted by practicing medical specialists to the regional health authorities. Coverage is considered complete as all claims must be submitted for reimbursement.

Relevant surgical procedures performed in hospitals were identified using the Classification of Surgical Procedures, which was developed by the Nordic Medico-Statistical Committee (NOMESCO), whereas TTI data were identified in the DNPR using the specific surgical code KDCA20 (TTI). For ENT practices, TTI data were identified in the NHSR using reimbursement codes 3009 (unilateral TTI) and 3109 (bilateral TTI).

Study Population and Design

The study population comprised 2,243,285 children aged 0–12 years, at one point registered as living in one of the five Danish administrative regions between 2003 and 2023. Data were collected from the Capital Region of Denmark, Region Zealand, the Region of Southern Denmark, the Central Denmark Region, and the North Denmark Region. The full cohort (hereinafter referred to as the calendar cohort) encompassed the total number and geographical distribution of procedures performed each year, as well as the at-risk population within specific time periods.

The study examined treatment trends by birth-year cohort, including a restricted analysis limited to children born in one of the five Danish administrative regions until age 2 or lost to follow-up. In the analysis of regional differences, three such birth cohorts were used to investigate temporal changes: children born in 2008–2009 (n = 127,679), 2017–2018 (n = 122,575), and 2021–2022 (n = 121,494). These three birth cohorts were selected because they represent three distinct time periods: children aged 0–2 years in 2010 when TTI procedures peaked, children aged 0–2 years immediately before the COVID-19 pandemic, and children aged 0–2 years immediately after the COVID-19 pandemic. For these three birth cohorts, we calculated overall cumulative incidence proportions and incidence rates stratified by the children’s region of residence.

In the analysis of the number of TTIs performed, we included all patients born between 2003 and 2018 with complete five-year follow-up data in the calendar cohort, covering the period from birth to age six.

Statistical Analyses

All analyses and graphical presentations were conducted using R statistical software.²³ The proportion of the population tubulated at age two and incidence rates were calculated from the survival functions derived from the Breslow cumulative hazard function using the survival package²⁴ with the rms package.²⁵ This allowed us to incorporate children moving between the 5 regions using truncation and follow the children until age 2 or censoring the children if lost to follow-up due to reaching 31^st of December 2023, unknown address for more than 30 days, emigration, or death. For the three birth cohorts, 2008–2009, 2017–2018, and 2021–2022, we lost 1,807, 1,867, and 51,306 children to follow-up before age 2 respectively, and we have a mean follow-up time among the children with no tubulation of 1.98, 1.98 and 1.73 years respectively. To preserve anonymity, all estimates were smoothed using moving averages.

Confidence intervals for ratios of proportions were computed using the rateratio-test package.²⁶

Ethical Approval

The Danish Data Protection Agency, the Danish Health Data Authority, and Statistics Denmark approved the study. Ethical approval and informed consent were not required, as the study was based on de-identified registry data encrypted and stored by Statistics Denmark. Reporting followed the STROBE guidelines. From 1 January 2003 to 31 December 2023, a total of 704,013 unilateral or bilateral TTI procedures were identified. They were distributed among 371,818 unique children aged 0–12 years. A total of 96.6% (n=679,756) of procedures were recorded in the NHSR, thus performed in ENT practices. The remaining 3.4% (n=24,257) were recorded in the DNPR and performed at public (99.8%) or private (0.2%) hospitals. Since the level of activity in private hospitals remained negligible throughout the observation period (n=485), their data were combined with those of public hospitals. Among the children receiving TTI, 57.8% were male and 42.2% were female. Among the children who had the surgery performed, 53.9% had it performed once, 24.5% twice, 11.2% three times, and 7.8% four or more times.

Temporal Trends and COVID-19

The total annual TTI number peaked in 2010 at 40,826 procedures per year and declined steadily to 29,846 by 2019. The steep decline seen in the total number of TTI procedures performed in 2020 and 2021 can be attributed to the COVID-19 pandemic (Figure 1). In 2022, the total annual number of TTIs returned to pre-COVID-19 levels and increased further in 2023, reaching 35,024 procedures. We calculated the proportion of the calendar cohort (aged 0–12 years) who received at least one TTI during a given year. In 2019, 25,555 children out of a total population of 821,337 received at least one TTI (3.11%), and, in 2023, 29,647 children out of a total population of 807,513 received at least one TTI (3.67%). This increase in proportion corresponds to a proportion ratio of 1.18 (95% CI [1.16–1.20]). In hospitals, 1078 children received a TTI in 2019 (0.13%), and 1561 children (0.19%) received one in 2023. This increase corresponds to a proportion ratio of 1.47 (95% CI [1.36–1.59]).

Figure 1.

(A) Number of total yearly TTI procedures among children aged 0–12, from 2003 to 2023. (B) The yearly proportion of children aged 0–12 years who received at least one TTI, from 2003 to 2023.

Age at First TTI Procedure and Regional Differences

Over the entire observation period (2003–2023), the median age at first TTI for children tubulated before age 12 across regions was 18.1 months. However, substantial regional differences were observed: 19.7 months in the Capital Region of Denmark, 18.0 months in Region Zealand, 18.1 months in the Region of Southern Denmark, 17.4 months in the Central Denmark Region, and 16.4 months in the North Denmark Region.

To explore whether the regional differences were consistent over time or exhibited temporal variation, we analyzed three birth cohorts spanning the entire observation period: children born in 2008–2009, 2017–2018, and 2021–2022, respectively. The proportion of children who had undergone at least one TTI by age two varied substantially across regions and over time (Table 1). By 2021–2022, proportions had declined in the Capital Region (12.6 per 1000) and Region Zealand (18.6 per 1000), whereas proportions had increased in the Central Denmark Region (24.6 per 1000) and the Region of Southern Denmark (20.7 per 1000). Figure 2 illustrates the incidence rates of first TTIs across the three birth cohorts (2008–2009, 2017–2018, and 2021–2022) stratified by region. In all regions, the incidence peaked at age one and declined thereafter. Incidence rates generally declined over time, having peaked in 2009–2010. However, in 2021–2022, a modest increase was observed in the North Denmark Region and the Central Denmark Region. Throughout the study period and across all three birth cohorts, incidence rates were highest in the North Denmark Region and lowest in the Capital Region of Denmark, reflecting persistent geographical variation.

Table 1.

Proportion of Children Having at Least One TTI Before Age 2, Grouped by Region for Three Birth Cohorts: Children Born in 2008–2009, Children Born in 2017–2018, and Children Born in 2021–2022

Region	2008–2009 Cohort	2017-–2018 Cohort	2021–2022 Cohort
Capital Region of Denmark	15.4	11.3	12.6
Central Denmark Region	20.4	21.6	24.0
North Denmark Region	25.9	20.4	24.6
Region Zealand	21.0	16.3	18.6
Region of Southern Denmark	18.1	16.5	20.7
Overall	19.0	16.2	18.8

Figure 2.

Incidence rates (per 1000 children per year) by region for the three birth cohorts: born in 2008–2009 (baseline), born in 2017–2018 (pre-COVID-19), and born in 2021–2022 (post-COVID-19).

TTI During the COVID-19 Pandemic

Figure 3 shows the monthly number of TTI procedures performed from January 2019 to December 2023.

Figure 3.

Monthly number of TTIs from January 2019 (pre-COVID-19) to December 2023, with the COVID lockdown period shaded.

Prior to the onset of the COVID-19 pandemic, the monthly number of TTI procedures typically ranged from 2500 to 3500, exhibiting clear seasonal variation, with lower volumes during summer, notably in July, likely attributed to holidays and subsequent reduced demand. This seasonal pattern persisted during the lockdown, albeit at lower overall levels.

A sharp decline occurred in March 2020, following the first nationwide lockdown in Denmark. Throughout 2020 and most of 2021, monthly procedure volumes remained markedly reduced, ranging from 1000 to 1500 procedures. Activity began to recover in late 2021 and into 2022, with counts occasionally approaching pre-pandemic levels, although intermittent declines persisted.

Discussion

The main finding of this large nationwide register-based study was a substantial decline in the total annual incidence of TTIs among Danish children since 2012, with a further reduction during the COVID-19 pandemic and a post-pandemic increase in 2023.

The present study expands on the earlier Danish register data reported by Djurhuus et al, who observed a consistent increase in TTI incidence rates between 1997 and 2010.¹¹

Similar findings have been observed in other Nordic populations, albeit with considerable variation. A Finnish study reported an increase in TTI rates from 97 to 147 per 10,000 children between 1999 and 2005,²⁷ while a Norwegian study reported relatively stable, yet regionally heterogeneous rates.¹² Another Finnish study found that TTI surgery was performed 2–3 times more frequently in Finland than in Norway, despite comparable OM prevalence, suggesting differences in clinical thresholds and practice traditions.¹² Substantial regional variation within Finland has also been reported, partly attributed to differences in the availability of ENT surgeons and variations in local surgical practice.²⁷ The observed variation in TTI rates across Nordic countries may reflect differences in access to ENT specialists, parental expectations, reimbursement models, healthcare system organization, childcare arrangements, and parental sick leave policies. Other research has indicated that international and regional variation in TTI rates often exceeds what can be explained alone by epidemiology.¹⁴

Hence, the regional variation observed in our study, with higher incidence rates in the North Denmark Region and Central Denmark Region and consistently lower rates in the Capital Region of Denmark, aligns with findings from other Nordic countries. Our findings indicate that regional differences cannot be explained solely by variations in disease burden or access to specialist care. In fact, the highest density of ENT specialists per inhabitant is found in the Capital Region of Denmark, where TTI rates are lowest. This suggests that, in addition to disease burden and parental preferences, local clinical culture, practice preferences, and referral behavior can play a role.¹²

The overall decline in TTI incidence between 2010 and 2019 is unlikely to be purely demographic, as the annual proportion of our study population undergoing tube insertion declines. We hypothesize the reductions instead might also reflect two converging factors: a 2015 national guideline promoting more restrictive, watchful-waiting-based indications for surgery,²⁸ and a reduced burden of pneumococcal-related otitis media following introduction of pneumococcal conjugate vaccines (PCV7 in 2007, PCV13 in 2010), consistent with the nationwide decline in tube insertion reported after PCV13.²⁹

The downward trend in TTIs observed between 2010 and 2019 was abruptly disrupted by the COVID-19 pandemic, which led to a pronounced decline from March 2020 to the spring of 2022. Thereafter, activity resumed markedly, approaching levels comparable to those observed before the long-term decline of the early 2010s.

The sharp reduction observed during the pandemic is consistent with international observations. In the Netherlands, a 63% reduction in OM incidence was observed during the pandemic; this was attributed to reduced respiratory virus transmission during lockdowns and school closures.³⁰ Similar findings were reported in the United States, where TTI rates dropped distinctly in 2020, and the median age at surgery increased, suggesting that younger children were more often managed medically or had their procedures postponed.³¹ In Canada, bilateral myringotomy and TTI rates decreased by 64% during the pandemic; this reduction is not explained by surgical capacity, but rather by fewer upper respiratory infections.³² Likewise, in Finland, a national registry study found that infection-related otolaryngological procedures, such as TTIs, declined more sharply than non-infection-related surgeries and remained at low levels throughout 2021. Collectively, these findings suggest that the decrease observed in our data reflects a true reduction in disease incidence rather than the effects of pandemic-related measures, such as lockdowns and reduced contact with the healthcare system for non-COVID-19 conditions.

The modest increase observed shortly after the pandemic likely reflects both a catch-up effect following deferred care and a genuine rise in incidence, as children were again exposed to respiratory infections after restrictions were lifted. During the pandemic, daycare institutions were closed nationwide in Denmark from 11 March 2020, with emergency childcare available exclusively for essential workers. Daycare reopened in April under strict hygiene and distancing measures, which were gradually eased and fully lifted by September 2021.

A notable feature of our dataset is that the rebound phenomenon occurred in both ENT practices and hospitals. To our knowledge, no previous studies have documented a comparable parallel development across healthcare sectors. This supports that the observed rebound was not solely resulting from local organizational prioritization but reflects a genuine epidemiological phenomenon.

Strengths and Limitations

This study has several important strengths. First, it is based on nationwide, population-based administrative registers providing a large and highly representative dataset without selection bias. Second, the unrestricted access to healthcare in Denmark reduces socioeconomic bias in access to treatment and enhances the generalizability of the findings. Third, the use of unique personal identification numbers enables accurate linkage across registers, ensuring precise identification of patients and procedures across healthcare sectors. Finally, the administrative data undergo continuous quality assurance, and reimbursement in primary care is monitored, which ensures a high degree of completeness.

This study also has several limitations. Most importantly, the administrative data used were not collected for research purposes, and clinical details such as severity of disease, duration of symptoms, prior treatment attempts or specific indications for TTIs are not available. As a result, variation in clinical practice patterns or patient characteristics cannot be fully explained. Moreover, there is a risk of misclassification or coding errors. Although the overall data quality is considered high, minor inaccuracies may affect estimates. Finally, the exact date of the procedure is not always available in the primary care register, as reimbursement claims may be submitted with delays of up to one month.

Overall, the data are regarded as high quality and well suited for population-based analyses. Nevertheless, the results should be interpreted with appropriate caution in light of these methodological limitations.

Conclusions

This nationwide register-based study found a substantial decline in TTIs among Danish children over the past decade. This was accompanied by an additional reduction during the COVID-19 pandemic, followed by a post-pandemic rebound. Regional differences in incidence and age at first TTI were similar to observations in other Nordic Countries. While registry data cannot be used to evaluate clinical indications or outcomes, the findings indicate shifting TTI practice patterns. It is, however, still too early to determine the long-term pandemic effects.

Funding Statement

The Fund for Professional Development of Medical Specialist Practice.

Data Sharing Statement

The data used in this study were obtained from the Danish Health Data Authority and Statistics Denmark and are not publicly available due to data protection regulations. Researchers may apply for access through the Danish Health Data Authority (Sundhedsdatastyrelsen) and Statistics Denmark (Danmarks Statistik).

Disclosure

The authors report no conflicts of interest in this work.