Background: Children and adolescents extensively use the internet in their daily lives, often seeking information related to health and well-being. In modern society, the volume of health information available in digital environments is constantly increasing. This includes both reliable and misleading content, making it challenging to assess trustworthiness. Digital health literacy is essential for navigating the digital information ecosystem, protecting oneself from misinformation, and making informed health decisions. Objective: This representative study aims to examine the digital health literacy of children and adolescents in Germany and its association with sociodemographic factors. Methods: A cross-sectional study design with face-to-face interviews was utilized to collect data from 1448 children and adolescents aged 9-18 years in Germany between October and November 2022. Digital health literacy was assessed using an adapted and translated version of the Digital Health Literacy Instrument (DHLI), which comprises 7 subscales: operational skills, navigation skills, information searching, self-generated content, evaluating reliability, protecting privacy, and determining relevance. Bivariate and binary logistic regression analyses were conducted to examine associations between digital health literacy subscales and sociodemographic characteristics (sex, age, migration background, school type, and perceived family affluence). Results: The study found that 419 out of 1362 (30.76%) children and adolescents had a problematic level of digital health literacy, while 63 out of 1362 (4.63%) had an inadequate level. Overall, the least difficulties were observed in operational skills and determining relevance, whereas the greatest challenges were related to protecting privacy and navigation skills. Age was significantly associated with 6 of the 7 subscales (excluding protecting privacy), with younger children (9-11 years) facing a higher risk of limited skills (operational skills: odds ratio [OR] 5.42, P=.002; navigation skills: OR 4.76, P<.001; information searching: OR 4.68, P<.001; adding self-generated content: OR 7.03, P<.001; evaluating reliability: OR 3.82, P<.001; and determining relevance: OR 4.76.42, P<.001). Migration background was associated with fewer limited digital health literacy skills, while low perceived family affluence was associated with more limited skills. In the subscales of information searching, self-generated content, and evaluating information reliability, a lower risk of limited skills was observed among those with a 2-sided migration background (information searching: OR 0.62, P=.02; adding self-generated content: OR 0.30, P=.003; and evaluating reliability: OR 0.66, P=.03). By contrast, a higher risk was found among those with low perceived family affluence, including in the subscale of determining relevance (information searching: OR 2.18, P<.001; adding self-generated content: OR 1.77, P=.01; evaluating reliability: OR 1.67, P<.001; and determining relevance: OR 1.58, P<.001). Although school type was not associated with any dimension, sex was linked to operational skills, with females having an increased risk of limited skills (OR 1.58, P=.03). Conclusions: The results highlight a strong need for interventions to improve digital health literacy among children and adolescents, particularly in protecting privacy, navigation skills, and evaluating the reliability of health information. Effective interventions should be tailored to address the varying needs associated with age, migration background, and family affluence.