Varicella (chickenpox) is a ubiquitous childhood disease caused by varicella zoster virus (VZV)—a DNA virus from Herpesviridae family—that typically manifests by fever, malaise and vesicular rash.1 2 3 4 5 , 6 7 4
The burden and epidemiology of varicella differ between countries due to various reporting and surveillance systems and health care seeking patterns.8 9–13 9–18 17
Few countries implement routine varicella immunization in childhood, including Norway, where varicella vaccines have been available since 1996, but at present are recommended only for specific risk groups.8 , 19 20 21 , 22 23
To this end, we aimed to characterize the health care burden of varicella in Norway by estimating the annual incidence of primary health care consultations, hospital contacts and deaths in prevaccine era, to assess the need for potential vaccine introduction. This is the first study assessing the burden and epidemiology of varicella in Norway at the national level, thereby providing evidence for national and international vaccine policy decisions.
MATERIALS AND METHODS
Study Design
This was a retrospective national study that assessed the use of health care resources by patients with varicella diagnosis and mortality due to varicella disease in the whole population of Norway during 2008–2014. We extracted individual patient data from several national health registries: the Norwegian Immunization Registry (varicella vaccinations),24 Registry (varicella-coded hospital contacts), the Cause of Death Registry (varicella-coded deaths) and the Norwegian Surveillance System for Communicable Diseases (viral infections of CNS).25 https://links.lww.com/INF/C747
Data Analysis
We calculated age- and sex-specific proportions (%) of patients vaccinated against varicella and incidence of varicella-related health care contacts in primary and hospital care per 100,000 population by using the numbers of patients with varicella diagnoses registered at the first contact during 2008–2014. Health care contacts were considered to be related to varicella regardless of which diagnostic field varicella diagnostic codes were assigned to. The age- and sex-specific population data by year were obtained from Statistics Norway.26
Patients with varicella-related hospital contacts were classified by using the information available from the International Classification of Diseases, 10th edition (ICD-10) codes on accompanying nonvaricella diagnoses. Two pediatricians reviewed all extracted ICD-10 codes and assessed the grouping of patients first by organ systems. Patients were subsequently assessed if they were presumably immunocompromised and/or had an underlying condition or comorbidity and were grouped accordingly (a full list of diagnostic groups applied is provided in the Table, Supplemental Digital Content 2, https://links.lww.com/INF/C748 27
To estimate varicella-associated mortality, we calculated crude and age- and sex-adjusted mortality rates per 100,000 by using the numbers of deaths with varicella-related diagnoses reported during 1996–2012. To calculate standardized mortality rate, we used age-standardized population data for Scandinavian countries from the World Health Organization.28
RESULTS
During the study period, a total of 76,168 patients with varicella-coded diagnoses consulted primary health care providers 100,096 times, which corresponds to 14,299 primary health care contacts made by 10,881 patients annually. Among these patients, 51.0% were males; children < 5 years of age accounted for 55.2%, and adults 20–39 years of age represented 9.6% (Fig. 1 ). In 96.1% (n = 73,168) of patients, varicella was reported as the primary diagnosis at their first contact. Most patients (75%) contacted General Practitioners and 78.5% made only 1 contact, whereas 2 and ≥ 3 contacts (range, 1–33) were registered for 15.7% and 5.9% of varicella patients, respectively. The annual number of varicella primary health care contacts was stable over the 7-year study period but a significant increase was observed annually in November, February and June.
FIGURE 1.: Proportion of varicella patients at first contact by a health care setting (hospital, primary healthcare) and age group, Norway, 2008–2014.
During 2008–2014, 2,526 patients were discharged from Norwegian hospitals with varicella-related ICD-10 codes listed on any discharge diagnosis. On average, 433 hospital contacts were made by 361 patients annually. Among these patients, 48.1% were inpatient, 49.9% were outpatient and 2.0% received ambulatory care. In 87.7% of the patients, only 1 hospital contact was reported, whereas 9.5% had 2 contacts and 3.5% had ≥ 3 contacts (range, 1–22). A higher proportion of hospital patients was male (54.3%), and median age was 5 years (Interquartile Range (IQR), 2–20 years). Children < 10 years of age accounted for 66.7% of all hospitalized patients, whereas 14% were adults 20–39 years of age. Varicella hospital contacts had a similar seasonal distribution as contacts in the primary health care. Overall, 99.5% of patients had varicella-specific codes listed on the first 3 discharge diagnoses, and 74.5% (95% CI: 72.8–76.2) had these codes assigned to primary diagnosis. Varicella as primary diagnosis was more often reported in children and outpatient cases, whereas varicella as secondary diagnosis was found in 25.5% (95% CI: 23.8–27.2) of patients and was more frequent among inpatient cases.
The median length of hospital stay among inpatient cases (n = 1284) was 3 days (IQR, 1–6). There were no differences by sex in the length of hospitalization but there was a significant association between age and length of hospital stay: the latter increased by 1 day for each 10 years of age (95%CI: 0.8–1.2). Patients with varicella-related complications (ICD-10 codes B010-B018) stayed in hospital 2.2 days longer (95%CI: 1.3–3.1) compared with patients without complications, whereas immunocompromised patients stayed almost a week longer (5.3 days; 95%CI: 3.8–6.8). Among hospitalized patients, 122 (4.8%) received intensive care, and their median age was 5 years (IQR: 2–17), similar to those not receiving such care. Median duration of intensive care was 2 hours (IQR: 0.8–41.5), but it significantly increased by 3.3 hours (95%CI: 2.5–4.0) per each year of patient’s age, reflecting an increasing severity of varicella with age. One half of intensive care patients had varicella-specific complications, but there was no difference in the duration of intensive care compared with patients without complications.
Among hospitalized patients, 91% did not have any comorbidity, and 9% had moderate-to-severe comorbidities according to the Charlson comorbidity index. The severity of comorbidities (Fig. 2 ) and the length of hospital stay increased with older age (Table 1 ). Twenty-nine percentage of hospitalized patients had varicella-related complications and 36% had other complications presumably related to varicella but without assigned varicella-specific codes. Among all patients with complications and comorbidities, between 30% and 80% were children 0–19 years of age, except for patients with HIV and AIDS. The most frequent complications were neurologic complications followed by conditions affecting a lower respiratory tract and skin. HIV and AIDS were reported in 0.4% of varicella patients, malignancies in 2.6%, autoimmune diseases in 4.4%, and 1.3% of the patients had undergone organ transplantations (Table, Supplemental Digital Content 3, https://links.lww.com/INF/C749
TABLE 1.: Frequency (%) of Comorbidities Among Hospitalized Patients (n = 2,526) With Varicella-specific ICD-10 Codes (B01-B019) on Discharge Diagnoses According to the Charlson Comorbidity Index by Age Group (Years) and Change in the Length of Hospital Stay (Days, 95% CI) Adjusted by Age and Sex, Norway, 2008–2014
FIGURE 2.: Age distribution (%) of hospitalized varicella patients by the Charlson comorbidity index, Norway 2008–2014.
To assess VZV-associated neurologic complications, we examined the 1997–2012 data about viral infections of CNS reported to the Norwegian Surveillance System for Communicable Diseases. The VZV was the third most frequent virus (16.3%) detected among 2,237 patients with reported viral infections of CNS preceded by enteroviruses (52.9%) and other Herpes viruses (Epstein Barr-virus and Herpes simplex viruses; 22.9%). Among patients with detected VZV in CNS, median age was 44 years (IQR, 25–72; Fig. 3 ). By linking these data to the hospital registry data for 2008–2012, we identified only 79.9% of varicella patients with confirmed VZV in CNS in the hospital registry , suggesting that remaining patients were assigned nonvaricella diagnostic codes.
FIGURE 3.: Proportion of VZV-caused and total viral infections of CNS by age group, Norway 1997–2012.
By linking primary care and hospital data, we estimated that during 2008–2014, 77,158 patients in Norway had 102,921 varicella-related contacts with the health care system, translating to 11,023 patients and 14,613 contacts on average annually. In addition, during the 7-year study period, we identified 552 (0.7%) patients who had both varicella and herpes zoster diagnoses, suggesting that these cases could have either developed herpes zoster after varicella infection or were misdiagnosed. Among patients with varicella, 96.8% contacted primary health care only, 1.9% contacted hospital only and 1.3% patients were in contact with both primary care and hospital sectors. The latter category had also a higher number of contacts per patient (on average 3 contacts per patient versus 1 contact in other groups).
The average annual incidence of varicella-related primary health care contacts was 221 patients (range, 164–274) per 100,000, and the highest incidence was in children at 1 year of age (2,654 patients per 100,000; Fig., Supplemental Digital Content 4, https://links.lww.com/INF/C750 https://links.lww.com/INF/C751 https://links.lww.com/INF/C752
During a 17-year period, 46 deaths were registered with varicella-related ICD-10 codes listed either as the underlying or contributing cause of death, corresponding to a crude mortality rate of 0.06 deaths per 100,000 (Table 2 ). The median age of deceased cases was 75.5 years (IQR, 38–83). Fifty-two percentage of deaths occurred in males. Half of the deceased had varicella listed as the underlying cause of death. Children < 18 years of age accounted for 11% (n = 5) of varicella deaths; of these 2 newborns had CVS that was listed as underlying cause of death. Among children who died of varicella, an underlying condition was reported only in 1 patient, suggesting that other deaths may have occurred in previously healthy individuals. Neurologic complications (varicella meningitis and encephalitis) accounted for 43.5% of all registered deaths and were predominant across all age groups. Varicella pneumonia was reported in 17% of deaths, all in adults.
TABLE 2.: Crude and Age- and Sex-adjusted Varicella Mortality per 100,000 Population Estimated by Using Varicella-specific ICD-10 Codes Listed Either as Contributing or Underlying Cause of Disease on Death Certificates, Norway 1996–2012
Among hospital varicella cases, 8 in-hospital deaths were reported during the 7-year study period corresponding to a case-fatality rate (CFR) of 0.3%. All deaths occurred in persons above 50 years of age, except for a single pediatric death. Additional 5 deaths (including 1 pediatric death) were identified within 30 days postdischarge that resulted in an overall CFR of 0.5% (n = 13). There were no sex differences in CFR. Among all 13 identified deaths (8 in-hospital and 5 postdischarge deaths), varicella was listed as the primary diagnosis in 8 cases, and underlying conditions were present in 8 patients.
During 2003–2012, a total of 4,021 persons were registered to have received 4,877 doses of varicella vaccine (on average 490 doses per year). One half of the vaccinated were children below age 1 year (50.3%), and 87.6% were children < 10 years of age. Linkage of immunization data with primary health care and hospital data indicated that a very few varicella patients were vaccinated: 0.2% (n = 126) of all primary health care patients and 0.5% (n = 12) of hospitalized patients.
DISCUSSION
This is the first registry -based study in Norway that employed a linkage of individual patient data across multiple national registries to quantify the health care burden of varicella. We used information from the national health registries because these represent nearly complete population-based databases, which represent unique data sources for conducting a comprehensive burden of disease assessment. As a very few patients in Norway were found to be vaccinated against varicella, our findings reflect varicella epidemiology compatible with a circulation of a wild-type VZV. Health care rates associated with medically attended varicella, and related mortality and in-hospital CFRs established in our study were within reported European ranges, but Norwegian primary health care rates were lower.17 8 , 17
The primary health care incidence of varicella was estimated in our study at 221 patients per 100,000, and the majority of varicella patients were in children < 5 years of age. Children with varicella in Norway were predominantly managed by primary care compared with adults who were more often hospitalized and stayed at hospital longer. Among all hospitalized varicella patients, 36% had complications, but this proportion was higher (54%) among children. Neurologic complications were most frequently reported, especially among adults, followed by the respiratory tract system and skin complications, which is in line with findings in other European countries.29
Our data suggest that 20% of patients with confirmed VZV infection in CNS might have been assigned other nonvaricella diagnoses that could not be captured in this study. Due to diagnostic challenges associated with detection of VZV in CNS,6 , 30 31
Our findings should be interpreted in light of potential limitations such as misclassification of varicella diagnosis and underreporting to health registries. Because a proportion of varicella cases with mild symptoms do not seek medical care, the true burden of the disease may be higher. In Belgium, about 40% of varicella patients were reported not to consult a physician.11
Even though Norwegian health registries are highly valued for their data quality, a disease-specific validation is the key element for the use of such data in research. The data from Norwegian Patient Registry were previously validated with varying results that were affected by a particular disease studied.32 , 33 registry varied between 98% and 35% across different regions in Norway and this could have impacted our results33 Registry .29
Completeness of Norwegian primary health care data, to our knowledge, has not been assessed, but because this information is used for reimbursement of health care providers we expect it to be high. It is however possible that a small proportion of varicella cases diagnosed in primary care may have been reported with nonspecific diagnoses such as “localized skin rash” (ICPC-2 code: S06). These cases therefore would not be captured in our study and this may have resulted in an underestimation of primary health care rates. As varicella vaccine is not included in the Norwegian immunization program, very few individuals were registered as vaccinated in the immunization registry . However, we cannot exclude that some vaccinations were not captured in the registry because a patient consent is required to report vaccinations not included in the national program. Similar to the situation in other countries,34 Registry .35 registry . Although hospital registry data are regularly updated against the Norwegian Population Registry , death dates do not always correspond with hospitalization dates, and thus it was challenging to estimate if the death was indeed related to varicella. Nevertheless, in the absence of other data source, using vital statistics data may be an acceptable way of quantifying varicella-associated deaths.
Despite above-mentioned limitations, this study demonstrates that for a rather easily recognizable disease such as varicella, the use of registry data may be an acceptable way to estimate health care contact rates, which potentially reflect the patterns of overall varicella incidence in the community. In conclusion, our study highlights the need for varicella vaccination in Norway, particularly for children and selected groups such as immunocompromised patients and nonimmune women of childbearing age. The latter 2 groups had the highest health care consumption, despite a lower incidence. Universal vaccination against varicella has reduced varicella hospitalizations and deaths in other countries,36 , 37
ETHICAL STATEMENT
The study has been approved by the Regional Committee for Medical and Health Research Ethics, Oslo, Norway, as well as an exemption from patient’s consent to use data from the registries.
ACKNOWLEDGMENTS
The authors acknowledge Beatriz Valcarcel Salamanca, Department for Infectious Diseases Epidemiology and Modeling, the Norwegian Institute of Public Health, for the assistance with statistical analysis, and Arild Osen, Department of Health Data Management and Analysis, the Norwegian Institute of Public Health for his valuable assistance during extraction, management and linkage of data from the registries.
REFERENCES
1. Heininger U, Seward JFVaricella. Lancet. 2006;368:1365–1376.
2. Johnson RWHerpes zoster and postherpetic neuralgia. Expert Rev Vaccines. 2010;9(3 Suppl):21–26.
3. Nardone A, de Ory F, Carton M, et al.The comparative sero-epidemiology of varicella zoster virus in 11 countries in the European region. Vaccine. 2007;25:7866–7872.
4. Bonanni P, Breuer J, Gershon A, et al.Varicella vaccination in Europe—taking the practical approach. BMC Med. 2009;7:26.
5. Grahn A, Studahl MVaricella-zoster virus infections of the central nervous system—prognosis, diagnostics and treatment. J Infect. 2015;71:281–293.
6. Quist-Paulsen E, Kran AM, Dunlop O, et al.Infectious encephalitis: a description of a Norwegian cohort. Scand J Infect Dis. 2013;45:179–185.
7. Gilden D, White T, Khmeleva N, et al.Prevalence and distribution of VZV in temporal arteries of patients with giant cell arteritis. Neurology. 2015;84:1948–1955.
8. European Centre for Disease Prevention and Control. Varicella Vaccination in the European Union. 2015.Stockholm: ECDC;
9. van Lier A, van Erp J, Donker GA, et al.Low varicella-related consultation rate in The Netherlands in primary care data. Vaccine. 2014;32:3517–3524.
10. de Melker H, Berbers G, Hahné S, et al.The epidemiology of varicella and herpes zoster in The Netherlands: implications for varicella zoster virus vaccination. Vaccine. 2006;24:3946–3952.
11. Bilcke J, Ogunjimi B, Marais C, et al.The health and economic burden of chickenpox and herpes zoster in Belgium. Epidemiol Infect. 2012;140:2096–2109.
12. Brisson M, Edmunds WJEpidemiology of varicella-zoster virus in England and Wales. J Med Virol. 2003;70 (Suppl 1):S9–14.
13. Socan M, Blaško MSurveillance of varicella and herpes zoster in Slovenia, 1996 – 2005. 2007;12:Euro surveillance: bulletin Europeen sur les maladies transmissibles = European communicable disease bulletin. 687.
14. MacIntyre CR, Chu CP, Burgess MAUse of hospitalization and pharmaceutical prescribing data to compare the prevaccination burden of varicella and herpes zoster in Australia. Epidemiol Infect. 2003;131:675–682.
15. Gil A, Oyagüez I, Carrasco P, et al.Epidemiology of primary varicella hospitalizations in Spain. Vaccine. 2001;20:295–298.
16. Korczynska MR, Rogalska JChickenpox in Poland in 2013. Przeglad epidemiologiczny. 2015;69:219–222, 345217.
17. Helmuth IG, Poulsen A, Suppli CH, et al.Varicella in Europe—a review of the epidemiology and experience with vaccination. Vaccine. 2015;33:2406–2413.
18. Hobbelen PH, Stowe J, Amirthalingam G, et al.The burden of hospitalisation for varicella and herpes zoster in England from 2004 to 2013. J Infect. 2016;73:241–253.
19. Norwegian Institute of Public Health. Varicella- and herpes zoster vaccination—guidelines for healthcare professionals (Varicella- og herpes zostervaksinasjon - veileder for helsepersonell). 2016. Norwegian Institute of Public Health; Available at:
www.fhi.no : Accessed August 22, 2016.
20. Poletti P, Melegaro A, Ajelli M, et al.Perspectives on the impact of varicella immunization on herpes zoster. A model-based evaluation from three European countries. PLoS One. 2013;8:e60732.
21. Hope-Simpson REThe nature of herpes zoster: a long-term study and a new hypothesis. Proc R Soc Med. 1965;58:9–20.
22. Ogunjimi B, Van Damme P, Beutels PHerpes zoster risk reduction through exposure to chickenpox patients: a systematic multidisciplinary review. PLoS One. 2013;8:e66485.
23. Lal H, Cunningham AL, Godeaux O, et alZOE-50 Study Group. Efficacy of an adjuvanted herpes zoster subunit vaccine in older adults. N Engl J Med. 2015;372:2087–2096.
24. Trogstad L, Ung G, Hagerup-Jenssen M, et alThe Norwegian immunisation register—SYSVAK. 2012;17.Euro surveillance: bulletin Europeen sur les maladies transmissibles = European communicable disease bulletin.
25. Løvlie A, Blystad H, Bruun TMSIS celebrates 40 years. Tidsskr Nor Laegeforen. 2015;135:2136–2138.
26. Statistics Norway. Population 2016. Available at:
http://ssb.no . Accessed August 25, 2016
27. Schmidt SA, Kahlert J, Vestergaard M, et al.Hospital-based herpes zoster diagnoses in Denmark: rate, patient characteristics, and all-cause mortality. BMC Infect Dis. 2016;16:99.
28. Ahmad OB, Boschi-Pinto C, Lopez AD, et alAge Standartization of Rates: A New WHO Standard. GPE Discussion Paper Series: No.31. 2001.Geneva, Switzerland: WHO;
29. Glode Helmuth I, Broccia MD, Glenthøj JP, et al.Children hospitalized with varicella in Denmark: sensitivity of the national patient register. Pediatr Infect Dis J. 2017;36:31–35.
30. Grahn A, Bergström T, Runesson J, et al.Varicella-zoster virus (VZV) DNA in serum of patients with VZV central nervous system infections. J Infect. 2016;73:254–260.
31. Amlie-Lefond C, Gilden DVaricella zoster virus: a common cause of stroke in children and adults. J Stroke Cerebrovasc Dis. 2016;25:1561–1569.
32. Bakken IJ, Nyland K, Halsteinli V, et al.Norsk pasientregister: Administrativ database med mange forskningsmuligheter [The Norwegian Patient
Registry ]. Norsk Epidemiologi. 2004;14:65–69.
33. Bakken IJ, Surén P, Håberg SE, et al.[The Norwegian patient register—an important source for research]. Tidsskr Nor Laegeforen. 2014;134:12–13.
34. Johansson LA, Björkenstam C, Westerling RUnexplained differences between hospital and mortality data indicated mistakes in death certification: an investigation of 1,094 deaths in Sweden during 1995. J Clin Epidemiol. 2009;62:1202–1209.
35. Pedersen AG, Ellingsen CLData quality in the causes of death
registry . Tidsskr Nor Laegeforen. 2015;135:768–770.
36. Marin M, Zhang JX, Seward JFNear elimination of varicella deaths in the US after implementation of the vaccination program. Pediatrics. 2011;128:214–220.
37. Siedler A, Rieck T, Tolksdorf KStrong Additional Effect of a Second Varicella Vaccine Dose in Children in Germany, 2009–2014. J Pediatr. 2016;173:202–206.
