JAMA Netw Open. 2025 Sep 2;8(9):e2529659. doi: 10.1001/jamanetworkopen.2025.29659.
ABSTRACT
IMPORTANCE: Preexisting health conditions complicate post-COVID condition diagnosis in children and adolescents, while the lack of standardized clinical phenotypes challenges its definition and epidemiological estimates. Prospective studies with robust methodologies are needed to minimize bias and confounders, yet they remain scarce in high-impact factor journals.
OBJECTIVE: To review, characterize, and assess the methodology used in studies examining post-COVID condition in children and pediatric populations in highly cited studies, which have greater visibility and are likely to be used in informing policy.
EVIDENCE REVIEW: Studies on PubMed and studies with high citations on Web of Science published through July 2024 were reviewed. Pediatric studies from journals with an impact factor of 5 or higher were included if they employed observational or risk-benefit designs. Studies were classified based on whether they included a SARS-CoV-2 test-negative control group or a non-test-negative group, which included studies with either no comparator or a different type of comparator. Joanna Briggs Institute and Risk of Bias in Nonrandomized Studies of Exposures tools assessed the risk of bias.
FINDINGS: Of 426 publications, 24 were analyzed; 9 studies (38%) used test-negative controls, while 15 studies (63%) did not (P < .001). Among studies with test-negative groups, 4 (44%) used prospective cohort designs vs 5 (33%) studies without a test-negative group. Demographic reporting of post-COVID condition cases varied. Sex was reported in 12 studies (50%), but the median (IQR) sample size was small (27 female patients [14-110]; 21 male patients [10-79]). Psychiatric history was often not reported (20 patients [83%]), as well as a lack of history of comorbidities (16 patients [67%]) or body mass index (15 patients [63%]). Among 9 test-negative control studies, 4 (44%) used matched control design, while only 1 (11%) accounted for confounders by sex-stratifying results.
CONCLUSIONS AND RELEVANCE: These findings highlight the need for rigorous study designs that minimize bias and confounding, ensuring a clearer definition of pediatric post-COVID condition and its sequelae. Employing true test-negative matched controls is essential for distinguishing common symptoms and assessing risk factors, while standardizing demographic reporting strengthens comparability and ensures consistency in patient selection.
PMID:41026493 | DOI:10.1001/jamanetworkopen.2025.29659
