Study population and setting
This prospective cohort sought to describe the transmission dynamics of children with COVID-19 in Catalonia, Spain. The study identified pediatric index cases (<16 years of age) from July 1st to October 31st 2020, from 120 pediatricians across 71 primary health centers and hospitals. Total and positive SARS-CoV-2 reverse-transcriptase polymerase chain reaction (RT-PCR) results were reported to the Catalan Agency for Quality and Health Assessment for the Catalan Epidemiological Surveillance Network and reference laboratories. The participating pediatricians collected clinical information for cases and their household contacts. During overloaded work days, pediatricians reported data from the first 5 positive cases per day. Contact tracing was conducted by the research group, and follow-up was conducted by the patient’s pediatrician or by telephone interview with the parents, using the questionnaire. The pediatric index case was defined when the child was the first infected household member. The child was considered a primary case when only the child tested positive in the household with no other household cases, or directionality could not be determined. The child was considered a secondary case if another household contact was symptomatic and tested positive via RT-PCR before the child. The secondary attack rate was calculated by the number of new infections among contacts divided by the total number of household contacts . Logistic regression was used to assess transmission risk factors between pediatric and adult index cases.
Summary of Main Findings
During the study, 417,578 tests were reported among children <16 years, and 26,665 positive cases (6.4%) were identified. Of these positive cases, 1,309 (4.9%) were recruited for the study, and 1,040 completed the clinical, epidemiological, and microbiological data. 47.2% (N=491/1040) of pediatric cases with full assessment data were asymptomatic, and most symptomatic cases were mild (52.8%, N=589/1040). Of these 1,040 children, 72.7% (N=756) were secondary to an adult case, while 5.0% (N=52) were secondary to another child. An additional 152 (14.6%) were primary cases, and the majority of those did not transmit infection to an additional household member (71.7%, N=109) with the remaining 28.3% unable to be determined (N=43). Among 80 children (7.7%), they were household index cases resulting in transmission. The secondary attack rate was estimated at 59.0%(N=167 among 283 contacts) of pediatric index cases vs. 67.6% (from 393 cases out of 581 contacts) among household contacts with adult index cases. No individual or environment risks increased the secondary attack rate in pediatric index cases, and the rate was lower during the school period (33.3%) than during the summer period (62.1%).
The study had a large number of cases identified using the pediatrician reporting, and had RT-PCR tests available for all child cases. They also were able to examine changes in transmission during the summer compared to during the school year because of their study time. Utilizing physicians with established relationships with these patients already, the final case classification is likely quite accurate compared to if it had been done by self-report or external researchers.
The major limitation was the non-response and lack of reporting on particularly heavy workload days. It is not clear how many SARS-CoV-2 infections were missed when only the first 5 cases in a day were reported on. While this reflects a challenge in human resources to conduct these studies, it highlights that this study could not enumerate all pediatric cases in Catalonia during this period. Additionally, the lack of response between positive identified cases and those who completed the questionnaire (N=26,665 vs. N=1,040) may reflect volunteer bias, with households who suspected transmission from their child’s primary illness more likely to participate. Alternatively, families with particularly sick children may have been less likely to participate if they were going to the hospital to take care of their child. Therefore, it is not clear how this biases the sample. Additionally, their definition of index cases was based on chronology of symptoms or testing, which may not have reflected the actual index case, but rather which family member was more concerned about COVID-19 and sought testing for the virus. Additionally, the analyses may have underestimated transmission from asymptomatic individuals, as symptomatic household members were identified as the index case from their symptoms.
This study describes transmission trends among household contacts due to pediatric index cases and secondary transmission among children. It highlights temporal trends for the school year as well, relevant to public health policymakers.
This review was posted on: 22 April 2021