In a recent study published in Plos Medicine, researchers investigated coronavirus disease 2019 (COVID-19)-associated morbidity in children, adolescents, and adults using healthcare data covering ~46% of the German population.
Preliminary studies have observed persisting, recurring, or new onset mental and somatic health issues beyond many individuals' four-week acute phase of COVID-19. The World Health Organization (WHO) and other health organizations made concerted efforts to gather data from a plethora of observational research studies to provide a pragmatic case definition for this new health phenomenon.
It is still uncertain whether unexplained health conditions present three months after the onset of SARS-CoV-2 infection holds for children and adolescents due to the lack of available data among younger age groups.
About the study
In the present study, researchers obtained data from German statutory health insurance companies for 1 January 2019 to 31 December, 2020. They identified individuals with reverse transcription-polymerase chain reaction (RT-PCR)-confirmed COVID-19, and the date of their first diagnosis served as the index date for both cohorts. The control cohort had individuals who matched 1:5 based on age, gender, and comorbidity-based propensity score.
The team monitored 96 prespecified incident morbidity outcomes documented in the second quarter at or after the index date, further aggregated into 13 symptom complexes and physical, mental, and physical/mental overlap domains. They selected 96 outcomes based on published data and clinical expertise; thus, it provided a sound basis for examining potential long-term sequelae of COVID-19 across multiple symptom complexes.
Moreover, the distributions of covariates in the COVID-19 and control cohorts were similar, which indicated successful balancing. The researchers used the Poisson regression model to estimate incidence rate ratios (IRRs) with 95% confidence intervals (95% CIs).
The study population comprised 157,134 individuals, of which 11,950 were children, including adolescents, and 145,184 were adults; also, 48.1% and 60.2% of these were females, respectively. While adolescents were in the zero to 11 years age group, adults were in the age group of 18 to 49 years. The mean follow-up time for children/adolescents and adults was 236 days and 254 days, respectively.
The IRR estimates were similar for age groups zero to 11 and 12 to 17, whereas IRs in children were consistently lower than those in adults. The specific outcomes with the highest IRR and IR of at least 1/100 person-years in the RT-PCR-confirmed COVID-19 cohort in children and adolescents were fatigue, cough, and throat/chest pain, with respective IRRs of 2.28, 1.74, and 1.72. In adults, these outcomes included smell and taste disturbances, fever, and dyspnea, with respective IRRs of 6.69, 3.33, and 2.88. For all 13 health outcomes, IRs per 1,000 person-years in the COVID-19 cohort were markedly higher in the control group than in children/adolescents and adults.
Notably, the authors noted a gradient in IRRs related to the COVID-19 severity, with the highest IRRs among individuals receiving intensive care in adults and children. The relative magnitude of increased documented COVID-19-related morbidity was comparable for the physical, mental, and physical/mental overlap domains. In the COVID-19 cohort, IRs were significantly higher in all 13 and 10 symptom complexes in adults and children/adolescents, respectively.
Although the estimated long-term sequelae of COVID-19 appear to be less pronounced in children and adolescents in absolute terms, high infection rates emphasize that post-COVID-19 cannot be dismissed among children and adolescents.
The study results for adults are in line with many previous international studies based on routine health data, which, in turn, provided indirect support for the validity of findings for children and adolescents. However, the authors acknowledged that their results were conservative for statistical significance. Since standard error estimators based on aggregate data could not be adjusted, the same individual was represented multiple times due to matching with replacement.
Consequently, the net effect of aggregate data on variance estimators of IRRs was unclear. Thus, confidence intervals and p-values estimates should also be interpreted with caution. Moreover, the study follow-up time was only three to six months within the post-COVID-19 phase.
Nevertheless, per the study findings, the long-term sequelae of COVID-19 cannot be completely ruled out in children and adolescents, though they appear to be less affected than adults. Outpatient and inpatient COVID-19 diagnosis was associated with higher long-term demand for healthcare services for a broad set of outcomes more than three months after confirmed SARS-CoV-2 infection. For instance, in children and adolescents, the IRR of new-onset mental health problems during follow-up was higher than in adults, and the opposite happened for the pulmonary symptom complex.
To conclude, the researchers emphasized the need for more detailed population-based studies with extended follow-ups to confirm their results for children and adolescents and their impact on healthcare systems.