Evaluating the Impact of Moderate and Severe Varicella Infections on Danish Children Under 18 Years Old: A Comprehensive Analysis From 2015 to 2023

Abstract

Background:

Varicella (chickenpox) caused by varicella-zoster virus (VZV) is a highly contagious pediatric disease. While it typically manifests as a mild disease, complications leading to hospitalization are not uncommon. Knowing the current disease burden, seasonality and risk groups is pivotal for evidence-based decisions on the introduction of a varicella vaccine.

Methods:

Using the Danish National Patient Register and medical helpline records from the Capital Region, we estimated the burden of VZV infections among children younger than 18 years in Denmark from 2015 to June 2023.

Results:

We identified 14,443 cases with annual incidence rates of 325/100,000 children for mild-to-moderate varicella cases (helpline calls), 35/100,000 for moderate-to-severe cases (outpatients) and 10/100,000 for severe cases (hospitalized). Mild cases were most prevalent in children 1–4 years old, while moderate-to-severe cases were most prevalent among infants <1 year old. Among hospitalized cases, 23.2% had underlying diseases and 47.3% experienced complications. In April 2022, rapid surge in all types of VZV cases occurred, where admissions surpassed the prepandemic level.

Conclusion:

This study documents that varicella is highly prevalent among Danish young children. Mild-to-moderately severe cases strain healthcare, causing increased helpline calls and posing a persistent hospitalization risk, especially for infants and children with underlying conditions.

Varicella also known as chickenpox is a common pediatric disease caused by varicella-zoster virus (VZV). It manifests as an itchy rash featuring vesicular blisters, often accompanied by mild-to-moderate fever and general malaise.¹ VZV can assume a latent state within nerve ganglia after the primary infection and can be reactivated as herpes zoster (shingles) many years after primary infection.² VZV is one of the most contagious pediatric diseases and approximately 90% of children contract the virus before reaching the age of 10.³ Immunity against the disease is considered lifelong.⁴

In Denmark, varicella does not attain the status of a notifiable disease, and the disease is not part of the European surveillance portfolio managed by the European Centre for Disease Prevention and Control. Consequently, there is no systematic surveillance system in place in Denmark to monitor the prevalence of varicella, rendering the precise burden of varicella disease in Denmark elusive.⁵ The annual incidence of varicella is estimated to be equivalent to the annual birth cohort, approximately 60,000 per year in Denmark, where most cases are children.^1,6,7 While most varicella cases exhibit a mild and benign course, complications or hospitalizations are not infrequent.⁸

Varicella is preventable by a live-attenuated vaccine. The varicella vaccination has demonstrated both safety and efficacy in impeding the onset of varicella, garnering the endorsement from the World Health Organization. The World Health Organization advocates for the consideration of varicella vaccines in countries where the disease exerts a substantial impact on public health and where there exists a potential for attaining high vaccination coverage, over 80%.⁹ Several countries have already successfully incorporated the varicella vaccine into their national immunization programs.^10,11 In Denmark, varicella vaccines are currently procurable through the private market necessitating out-of-pocket expenditures.¹²

European studies, where the vaccine is part of national immunization programs, show promising findings in the reduction of the incidence of varicella and related hospitalizations.^10,11,13 A US study investigating the impact of the vaccine from 1993 to 2019 found a 97% reduction of varicella hospitalization for children.¹⁴ Noteworthy in Denmark, as in many other countries, concerns remain about the varicella vaccine. Specifically, there is apprehension about a potential age shift toward older age groups, which could lead to an increase in disease severity and elevated risk of herpes zoster later in life.¹⁰ Although current vaccine studies show promising benefits and encouraging evidence, its long-term effects, including potential drawbacks, remain uncertain and require further investigation to draw conclusive results.^4,15

With the approval of the varicella vaccine to be used in Denmark in 2019, and the recent documentation of increased uptake, despite the need for private purchase,¹⁶ there is a need for an updated study investigating the burden and seasonality of varicella among children in Denmark. This study aims to provide valid disease burden estimates of moderate and severe varicella infections among children under 18 years of age in Denmark. These estimates, along with identifying risk groups and understanding the seasonality of varicella, are essential for the optimal timing of prevention strategies and potential adjustment of vaccination schedules. The study period, from 2015 to June 2023, is crucial for covering the impact of the introduction of the vaccine in the private market and accounting for changes in virus transmission dynamics due to national lockdowns imposed due to the COVID-19 pandemic.

METHODS

Study Design

The study is a retrospective nationwide population cohort study based on national healthcare registry databases.

Data Sources

All our data were obtained between January 1, 2015 and June 30, 2023. The data were collected through the Danish National Patient Register (DNPR). The DNPR contains nationwide longitudinal registration of administrative and clinical data on an individual level from all Danish hospitals and emergency rooms¹⁷ and was used to collect data on all varicella inpatient and outpatient visits with both primary and secondary varicella diagnoses. DNPR contains the start and end date of inpatient and outpatient contacts and discharge diagnoses. The diagnoses are based on “The International Classification of Disease,” 10^th revision (ICD-10) codes. It is possible to register one primary diagnosis, while there is no limit to the number of secondary diagnoses. We extracted all ICD-10 diagnoses in any diagnostic positions to access other clinical characteristics and underlying medical conditions related to the varicella diagnosis. DNPR provides an opportunity to identify children hospitalized with varicella infections and outpatients’ visits and to estimate the age-specific incidence of varicella infections, thereby assessing the burden of varicella hospitalizations and outpatients.

We obtained data on the number of the background population from “Statistics Denmark” to calculate the incidence rates.¹⁸

In addition to the DNPR, we utilized data from the Non-Emergency Medical Helpline (named 1813 helpline). Data from the 1813 helpline contain information on every telephone consultation from inhabitants in the Capital Region of Denmark to the helpline. The helpline is mostly used outside the local practitioner’s telephone hours, for injuries or if an individual suddenly becomes ill. The purpose of the helpline is to assess whether the patients should be referred to a hospital. The helpline is staffed by specialized physicians and nurses, who register the described symptoms and presumptive diagnosis of the patients into a database called Logis CAD using codes and free text.¹⁹ In collaboration with the data team from the Non-Emergency Medical Helpline Department, we extracted a dataset with all the relevant diagnoses and symptoms from the database.¹⁹

These data are used to identify generally milder cases with varicella not registered in the DNPR. These data enable us to assess the incidence of mild cases of varicella among children from the Capital Region of Denmark. The 1813 helpline data contain information on every received contact at the individual level. A case of varicella infections or a case with varicella-like symptoms in the 1813 data is judged and registered by a specialized physician or nurse based on the caller’s description of the symptoms and the expected clinical characteristics of varicella disease.

We used the Danish Register of Causes of Death, which includes all deaths in Denmark, to describe mortality due to varicella infections. The cause of death is classified according to ICD-10 classifications for natural deaths.²⁰

Study Population

We included all children <18 years old born and/or living in Denmark during the period January 1, 2015 to June 30, 2023. All Danish citizens receive a unique identification number called a CPR number upon birth or immigration into Denmark enabling us to use the entire country as a national cohort of individuals. We defined the index date of a varicella case as the first day of contact with either the hospital or the 1813 helpline. Mild cases were defined as cases identified through the 1813 helpline in the Capital region. Approximately 368,429 (32.0%) children from Denmark lived in the Capital Region in 2023.¹⁸ To identify the moderate-to-severe varicella episodes, we utilized the DNPR to capture all the outpatient contacts and hospitalizations of children diagnosed with varicella, identified with the ICD-10 codes: B01–B019 and P358A. Following the 2019 update of the DNPR, patients are no longer registered as outpatients or inpatients. Instead, the duration of each contact is registered along with information about the type of contact (physical, virtual, external contact, death or diagnosis recording). To define moderate cases, we apply the Danish standard definition of an outpatient, with a hospital admission with a duration of less than 12 hours.²¹ All outpatients’ visits are coded with an ICD-10 code after the contact with the physician from a Danish hospital or emergency room, even though the patient is not hospitalized. The moderate cases are also referred to as outpatient throughout this article. Severe cases were defined as patients hospitalized with a varicella-related ICD-10 code for 12 hours or more. Several admissions with varicella for the same patient were defined as the same hospitalization. Due to different data sources, cross-referencing mild cases with outpatient cases and hospitalized cases was not feasible. See Figure, Supplemental Digital Content 1, http://links.lww.com/INF/F790 for the flow chart diagram illustrating participant inclusion from the 2 data sources.

Definitions of Complications, Length of Admission and Seasonality Assessment

Complications related to varicella were identified by extracting all other discharge diagnoses in DNPR during the admission with varicella. The included complications were selected based on previous literature (Table 1).^4,22 Underlying diseases related to varicella were defined as diseases occurring before the hospitalization with varicella and were extracted up to 5 years before the hospitalization with varicella. All diagnoses, both primary and secondary in any diagnostic positions, were included (See Table, Supplemental Digital Content 2, http://links.lww.com/INF/F791). Since a previous study has highlighted varicella as a risk factor for arterial ischemic stroke (AIS),²³ we extracted diagnoses related to AIS during VZV admission and up to 12 months after the VZV-related admission. All the applied ICD-10 codes are listed in Table, Supplemental Digital Content 3, http://links.lww.com/INF/F792; Table, Supplemental Digital Content 4, http://links.lww.com/INF/F793; Table, Supplemental Digital Content 5, http://links.lww.com/INF/F794; and Table, Supplemental Digital Content 6, http://links.lww.com/INF/F795.

TABLE 1.

Types of Complications in Children With Moderate and Severe Varicella in Denmark 2015–June 2023

	Outpatient Cases, n (%)	Hospitalized Cases, n (%)*
All patients	3509	871
Any complications	667 (19.0%)	412 (47.3%)
Complications of the central nervous system	256 (7.3%)	140 (16.1%)
Encephalitis, meningitis and myelitis	77 (2.2%)	57 (6.5%)
Ataxia	12 (0.3%)	6 (0.6%)
Afebrile seizures and status epilepticus	10 (0.3%)	6 (0.6%)
Febrile seizures	139 (4.0%)	60 (6.8%)
Stroke and cerebral vasculitis	10 (0.3%)	7 (0.8%)
Other	4 (0.1%)	4 (4.6%)
Complications of the muscular-skeletal system	20 (0.6%)	10 (1.1%)
Septic arthritis, osteomyelitis, synovitis, tenosynovitis and myositis	8 (0.2%)	5 (5.7%)
Aseptic arthritis	12 (0.3%)	5 (5.7%)
Complications of the blood, circulatory system and heart	27 (0.8%)	20 (2.3%)
Sepsis	15 (0.4%)	12 (1.4%)
Endocarditis and myocarditis	0 (0%)	0 (0%)
Other	12 (0.3%)	8 (0.9%)
Complications of skin and soft tissue	207 (5.9%)	77 (8.8%)
Infection of skin and soft tissue	185 (5.3%)	71 (8.2%)
Other	22 (0.6%)	6 (0.6%)
Complications of the lower respiratory tract	100 (2.8%)	45 (5.2%)
Pneumonia	44 (1.3%)	19 (2.2%)
Asthma	25 (0.7%)	10 (1.1%)
Bronchitis and bronchiolitis	31 (0.9%)	16 (1.8%)
Complications of the upper respiratory tract	121 (3.4%)	25 (2.9%)
Complications of the eye	19 (0.5%)	4 (0.5%)
Complications of the gastrointestinal system	47 (1.3%)	12 (1.5%)
Gastroenteritis	40 (1.1%)	8 (0.9%)
Stomatitis	1 (0.02%)	0 (0%)
Appendicitis	6 (0.2%)	4 (0.5%)
Unspecified bacterial infections	203 (5.8%)	51 (5.9%)
Complications of the urinary and kidney system	29 (0.8%)	6 (0.7%)
Glomerular diseases	11 (0.3%)	2 (0.2%)
Cystitis and urinary tract infection	6 (0.2%)	1 (0.1%)
Nephritis	5 (0.1%)	1 (0.1%)
Infections of genitals	7 (0.2%)	2 (0.2%)
Other
Dehydration	101 (2.9%)	34 (3.9%)

^*Mild cases are defined as cases registered through the 1813 nonemergency helpline. Mild cases are defined based on varicella-like symptoms, therefore related diagnoses and primary diagnoses were unavailable (only data from the Capital Region).

Length of hospitalizations was calculated as days between the first admission with varicella and the last discharge day.

Seasonality was defined as systematic periodic fluctuations in disease incidence that occur over the course of a year. These fluctuations were characterized by their magnitude, timing and duration. The time of the seasonal peak was defined as the position of the maximum point on the seasonal curve.²⁴

Statistical Analysis

For the descriptive analysis, we divided the varicella cases into 3 severity categories. The distributions of the continuous variables were skewed; therefore, we reported medians and interquartile rates (IQR). Categorical data were reported as frequency counts and percentages for each category. Age groups were: 0 years, 1–4 years, 5–9 years and 10–17 years.

Incidence rates of mild, moderate and severe varicella were estimated per 100,000 persons in the age group. For the mild cases, which only include cases from the Capital Region of Denmark, the denominator for the incidence rates contains the number of children in each specific age group living in the capital region during the period.

To investigate the determinants for severe hospitalization with varicella, a logistic regression analysis was performed to compare moderate and severe cases. The analyzed determinants for severe varicella were young age (dichotomized in age ≥1 or <1 year) and underlying diseases. Age under 1 year was specifically selected due to its association with a higher risk of severe disease outcomes.³ For the analysis, the underlying diseases were categorized into groups including cancer, autoimmune disease, central nervous system (CNS) disorders, lung disease, heart disease, other immunosuppressive conditions and a comprehensive category called “underlying diseases in general” encompassing all of them. For the adjusted analysis, we adjusted each analyzed covariate for sex and age and analyzed them independently. All statistical analyses were carried out using R statistical software version 4.2.2 (R Foundation for Statistical Computing, Vienna, Austria).²⁵

RESULTS

Demography and Seasonality of the Varicella Cases

Between 2015 and June 2023, a total of 10,063 cases of mild VZV infection were identified among Danish children living in the Capital Region, which covers approximately 32.0% of the children <18 years old in Denmark.¹⁸ An average of 1184 varicella-related cases were in contact with the helpline each year, ranging from 574 in 2021 to 1678 cases in 2015.

Through the DNPR, covering the entire Danish population, we identified 3509 outpatient visits (moderate cases) and 871 hospitalizations (severe cases) with a varicella-specific ICD-10 code among children from 2015 to June 2023. Through the Danish Register of Causes of Death, covering all deaths among citizens in Denmark, no deaths were reported among children during our study period.

Most varicella cases occurred in young children 1–4 years old, constituting 66.0%, 59.2% and 53.2% of mild, moderate and severe cases (Table 2). The median age was 2 years (IQR: 1–4). The baseline characteristics are presented in Table 2. For the severe cases, the median length of admission was 1 day (IQR: 1–4).

TABLE 2.

Baseline Characteristics of Mild, Moderate and Severe Varicella Infections Among Danish Children, Between 2015 and Mid-2023

	Mild Cases*	Moderate Cases†	Severe Cases‡
Number of contacts/hospitalizations, n	10,063	3509	871
Males, n (%)	5100 (50.7%)	1872 (53.3%)	478 (54.9%)
Age, median (IQR)	2 (1–4)	2 (1–4)	2 (1–5)
Age groups
0 yr	1414 (14.0%)	613 (17.5%)	185 (21.2%)
1–4 yr	6638 (66.0%)	2079 (59.2%)	463 (53.2%)
5–9 yr	1710 (17.0%)	594 (16.9%)	151 (17.3%)
10–17 yr	301 (3.0%)	223 (6.4%)	72 (8.3%)
Varicella as primary diagnosis	-	2880 (82.1%)	458 (52.6%)
Patients with underlying diseases§, n (%)	-	362 (10.3%)	202 (23.2%)
Patients with complications¶, n (%)	-	667 (19.0%)	412 (47.3%)
Length of stay (d), median (IQR)	-	-	1 (1–4)

^*Mild cases are defined as cases registered through the 1813 nonemergency helpline. Mild cases are defined based on varicella-like symptoms, therefore related diagnoses and primary diagnoses were unavailable (only data from the Capital Region).

^†Moderate cases are defined as patients registered as outpatients or inpatients for less than 12 hours.

^‡Severe cases are defined as patients registered for more than 12 hours.

^§Patients with underlying diseases are registered with an underlying medical ICD-10 diagnosis in the DNPR up to 5 years before the varicella diagnosis.

^¶Complications are defined as being registered with an ICD-10 diagnosis compatible with complications of varicella in DNPR at the time of hospitalization for varicella.

More mild cases were identified during the late winter months of February–May than the rest of the year, indicating a seasonal variation (Fig. 1A). The seasonal rises started in December, peaked the following April or May (the position of the maximum number of cases per month) and continued until the beginning of June. The seasonality of the moderate cases closely follows the same seasonality of the mild cases. However, the seasonal trend of the severe cases was not so clear, though most hospitalizations occurred during February, March and April. Coinciding with the COVID-19 pandemic, the number of all cases decreased from early 2020 to the end of 2021, and varicella cases almost disappeared. At the beginning of the spring 2022, a delayed seasonal peak was observed, lasting until June 2022 (Fig. 1A). In the first quarter of 2023, a premature peak for both the moderate and severe cases was observed.

FIGURE 1.

Occurrence of varicella infections among children <18 years of age, Denmark 2015–June 2023. A: Seasonal variation of mild, moderate and severe varicella infections per month. B: Age-specific incidence of mild varicella infections per quarter. C: Age-specific incidence of moderate varicella infections per quarter. D: Age-specific incidence of severe infections per quarter.

Incidence of Varicella Cases

The annual incidences were 325 per 100,000 for mild cases, 35 per 100,000 for moderate cases and 10 per 100,000 for severe cases. The age-specific annual incidence of mild cases was highest among the 1–4-year-old patients followed by the 0 years age group, respectively, 951.5 and 766.2 per 100,000 (Fig. 1B; Table, Supplemental Digital Content 6, http://links.lww.com/INF/F795). The highest incidences of moderate and severe cases were among the 0-year-old cases, respectively, 119 per 100,000 and 39 per 100,000, albeit closely followed by the incidence of the 1–4-year-old cases, 101 per 100.000 and 26 per 100.000 (Fig. 1C, D). A 54.6% decline in the total incidence of all varicella cases was observed between the second quarter of 2020 until the first quarter of 2022, compared to the pre-COVID period. In this period, the incidence ranged between 156–161/100,000 for mild cases, 15–21 for moderate cases and 3–5 for severe cases. In the first quarter of 2022 and until our data period ended in June 2023, the quarterly incidence for the 5–9 age group increased by 82%, 121% and 111% for mild, moderate and severe cases, respectively, compared to previous years (Fig. 1B, C).

Underlying Diseases and Complications Related to Varicella

Among the hospitalized children, 457 (52.5%) had varicella as their primary diagnosis, while for the outpatients, varicella was the primary diagnosis for 2880 cases (82.1%) (Table 2).

Among the hospitalized patients, 47.3% had complications and 1 in 5 (23.2%) had underlying disease (Tables 1 and 2; Table, Supplemental Digital Content 2, http://links.lww.com/INF/F791). For the outpatient cases, 19.0% had complications and 1 in 10 (10.3%) had underlying diseases. The most common complications for hospitalized cases were CNS related (31.3%), skin infections (18.8%) and unspecified bacterial infections (12.4%). In the outpatients, they were unspecified bacterial infection (21.4%), skin (19.0%) and CNS (16.5%) (Table, Supplemental Digital Content 2, http://links.lww.com/INF/F791).

Additionally, we identified 10 cases, among the hospitalized children with varicella, each concurrently diagnosed with an AIS-related diagnosis. One patient hospitalized with VZV was hospitalized with an AIS-related diagnosis 6 months later (data not shown).

Determinants for Severe Varicella

We performed a logistic regression analysis to investigate the determinants for severe varicella. We compared the moderate cases to the severe cases and explored possible determinants in crude and adjusted analysis. The determinants for being admitted with varicella (severe cases) were young age (<1 year of age) [odds ratio (OR):1.29, 95% confidence interval (95% CI): 1.07–1.55, P = 0.006], underlying disease in general (OR: 2.12, 95% CI: 1.74–2.58, P < 0.001), cancer (OR: 11.33, 95% CI: 5.63–22.76, P < 0.001), autoimmune disease (OR: 3.62, 95% CI: 2.17–6.02, P < 0.001), CNS infections (OR: 2.4, 95% CI: 1.63–3.53, P < 0.001), chronic lung disease (OR: 1.38, 95% CI: 1.02–1.88, P = 0.04) and chronic heart disease (OR: 1.49, 95% CI: 1.00–2.23, P = 0.04) (Table 3). In the adjusted analysis, the odds of being hospitalized were significant higher in children under 1 year of age (OR: 1.29, 95% CI: 1.07–1.55, P = 0.007) and in patients presenting with underlying disease in general (OR: 2.17, 95% CI: 1.77–2.64, P < 0.001), cancer (OR: 11.50, 95% CI: 5.71–23.17, P < 0.001), autoimmune diseases (OR: 3.64, 95% CI: 2.18–6.08, P < 0.001), CNS infections (OR: 2.44, 95% CI: 1.65–3.59, P < 0.001), chronic lung disease (OR: 1.42, 95% CI: 1.04–1.93, P < 0.025) and chronic heart disease (OR: 1.51, 95% CI: 1.02–2.25, P = 0.043).

TABLE 3.

Logistic Regression Analysis: Determinants for Severe Varicella Infections Compared to Moderate Infections

	Admitted Patients, n = 871	Outpatients, n = 3509	Crude Analysis			Adjusted Analysis*
			OR	95% CI	P Value	OR	95% CI	P Value
Underlying disease	202 (23.2%)	362 (10.3%)	2.17	1.78–2.65	<0.01	2.12	1.82–2.71	<1e-04
Cancer	29 (3.3%)	10 (0.28%)	12.04	5.84–24.80	<0.01	12.14	5.88–25.72	<1e-04
Autoimmune disease	28 (3.2%)	30 (0.85%)	3.85	2.29–6.47	<0.01	3.85	2.28–6.50	<0.01
Other Immunosuppressive conditions	11 (1.26%)	28 (0.80%)	1.59	0.79–3.20	0.20	1.56	0.78–3.14	0.18
CNS†	42 (4.8%)	71 (2.0%)	2.45	1.66–3.62	<1e-04	2.48	1.68–3.67	<1e-04
Lung disease	58 (6.7%)	169 (4.8%)	1.41	1.03–1.92	0.03	1.44	1.06–1.97	0.02
Heart disease	34 (3.9%)	92 (2.6%)	1.52	1.02–2.28	0.04	1.53	1.03–2.29	0.04
Age <1 yr	185 (21.2%)	613 (17.5%)	1.28	1.06–1.54	0.01	1.28	1.06–1.53	0.01
Male gender	478 (54.9%)	1872 (53.3%)	1.06	0.92–1.23	0.42	1.06	0.92–1.23	0.44

^*Covariates were adjusted for age group and sex. Age <1 year was adjusted for sex, and sex was adjusted for age groups.

^†Central nervous system and neuromuscular disease.

DISCUSSION

We conducted a comprehensive assessment of the burden of VZV infection among children in Denmark from 2015 to June 2023. One of the most important findings of this study was the discovery of the disease burden among milder cases, which also enhances the reliability of the overall varicella disease burden in a Scandinavian country. However, many children experience mild disease and may not utilize the healthcare system or the medical helpline, therefore our burden estimates are likely conservative. A previous study estimated that approximately 33,123–40,842 Danish children visit their general practitioner every year, due to varicella.²⁶ In comparison with other European studies, our hospitalization rates align with these incidence estimates ranging between 9 and 75 per 100,000 before vaccination introduction.²⁶ An important consideration is that the cost-effectiveness of introducing the varicella vaccination in Denmark has recently been estimated to reduce varicella cases by 94%–96%, and varicella-related hospitalization burden by 93%–94%.²⁷ The exclusion of vaccination status for moderate and severe cases is a limitation of our study. This was not durable to the low vaccination coverage, with only about 3.1% of children in Denmark in 2023 having received the first dose of the vaccine.¹⁶

Another key finding in this study was the impact of the pandemic lockdown and reopening of the country on varicella transmission, which has not been demonstrated previously. Before the COVID-19 pandemic, mild and moderate cases of varicella circulated mostly during the spring, with an onset in the winter months. During 2020–2021, a significant reduction in varicella transmission occurred, with cases nearly absent until early 2022, attributional to the implemented societal restrictions and preventive measures. A delayed peak was observed in the spring 2022 postpandemic, lasting until June 2022, and in 2023, a peak occurred in March, April and May for moderate and severe cases, higher than in the prepandemic period. An increase in incidence after 2022 in the 5–9-year old was observed, probably because these children were not infected during the pandemic, at ages of 1–4 years.

We found that 47.3% of the severe cases were registered with varicella-related complications during their admission. Our estimate is somewhat lower compared to complication rates in an older Danish study (67.1%), a Swiss study (76%) and a German study (80%),^4,8,28 probably due to differences in the study design. Among the severe cases, the most common complications were related to CNS (16%), skin (9%) and unspecified bacterial infections (6%), which were comparable to findings in a previous Danish study and a Swiss study.^4,28 Among the hospitalized children, 22.4% had a hospital contact due to a pre-existing medical condition within the past 5 years, predominantly associated with the respiratory system, cardiovascular system and/or CNS. Beyond the immediate strain on the healthcare system explored in this study, varicella can lead to additional complications and long-term sequelae, and 0.4%–8% of children hospitalized for varicella experience long-term sequelae.^29–31 In our logistic regression analysis, age below 1 year and underlying diseases were identified as determinants for severe varicella. As anticipated, the likelihood of hospitalization significantly increased for children with a medical history of cancer, CNS disorders, lung disease, autoimmune disease and/or heart diseases. However, this study also revealed that most of the children experiencing moderate-to-severe varicella in this study were previously healthy.

Some limitations are encountered when applying register-based data to explore the burden of diseases. The registry data may contain information bias and lead to underestimation of the incidence due to misreporting in ICD-10 coding in clinical practice. The DNPR is considered a high-quality database, however, a previous Danish study found that the sensitivity of varicella-coded hospitalization in DNPR was 74%.³ Additionally, we were not able to confirm the cases molecularly through laboratory tests. Albeit it is standard procedure to laboratory-confirm the cases admitted to a hospital in Denmark. The cases in this study are only identified by the use of ICD-10 codes.

Furthermore, most minor complications and infections related to varicella are handled at home or by general practitioners and will never be registered in DNPR or the 1813 helpline system. The 1813 helpline data cover solely the Capital Region of Denmark. Also, the purpose of the 1813 helpline is to assess the severity of symptoms, rather than diagnosis, potentially leading to inaccuracies in the number of mild cases. Additionally, without cross-linking between the cases from 1813 calls and the hospital admission, some of the mild cases identified through 1813 calls might have been among those hospitalized with varicella in the Capital Region. There were no reported deaths from varicella in children in our study period. However, a previous Danish study estimated the mortality rate of varicella infections to be 0.0014 per 100,000 in children under 18 years of age.⁴

CONCLUSION

Our results documented a high incidence of varicella infections among Danish children and related considerable use of healthcare services. By using data from the 1813 helpline, we documented that even milder illness poses a considerable burden on the healthcare system with frequent calls to the helpline. The results of this study are highly relevant in the consideration and evaluation of implementing the varicella vaccine in the Danish national immunization program.

ACKNOWLEDGMENTS

We thank Stine Munck and Sanne Enok Lauridsen from Emergency Medical Services Copenhagen for providing us with data and knowledge regarding data collection and procedures.