**Background:** Between March 2020 and February 2021, Baja California, Mexico—a state bordering the United States—registered 46,118 confirmed COVID-19 cases with a mortality rate of 238.2 deaths per 100,000 residents. Due to limited testing capacity and a passive surveillance system that tested only 10% of mild cases and 100% of severe acute respiratory illnesses, the true population prevalence of SARS-CoV-2 infection was unknown. As of January 2022, Baja California had registered 116,870 confirmed cases and had the second-highest mortality rate nationally after Mexico City, with 11,451 confirmed deaths (320 deaths per 100,000 population). This study aimed to estimate seroprevalence and RT-PCR prevalence of SARS-CoV-2 infection in the three most populous cities prior to the scale-up of Mexico's national COVID-19 vaccination campaign, during which only 0.3% of Baja California residents had received a vaccine by February 2021.
**Methods:** A population-based household survey was conducted from February 1 to February 19, 2021, immediately following the state's second wave of COVID-19 cases. A probabilistic three-stage clustered sampling design was employed. The target sample size of 1,500 was calculated to estimate a 4% prevalence with 30% precision, 95% confidence limits, and a design effect of 1.9. Within each city, 33 Basic Geo-Statistic Areas (AGEBs) were selected with probability proportional to population size, then eight blocks per AGEB, and four households per block through systematic random sampling. One participant aged 5 years or older was randomly selected per household. Inclusion criteria were Spanish speakers residing in Baja California for at least six months. RT-PCR testing was performed on nasopharyngeal and oropharyngeal swabs targeting the SARS-CoV-2 N1 gene (positive if Ct ≤ 40) at the Baja California Public Health State Laboratory. IgG antibody testing was performed on dried blood spots from fingerstick samples using ELISA at the Broad Institute Serology Lab. An interviewer-administered questionnaire assessed knowledge, attitudes, and preventive practices. Statistical analyses used weighted estimates with inverse selection probability adjustments, non-response adjustment, and calibration to 2020 Census population data.
**Key Results:** Out of 2,898 households visited, 1,283 (44%) agreed to participate, and 1,126 of 1,267 randomly selected individuals (89%) consented (35% from Mexicali, 35% from Tijuana, 30% from Ensenada), representing a weighted population of 2.8 million residents. Mean age was 37 years (95% CI 35–40), 50% were female, and 53% had less than a high school education. The overall weighted prevalence of SARS-CoV-2 infection by RT-PCR was 7.8% (95% CI 5.5–11.0), with Ensenada having the highest prevalence at 22.2% (95% CI 15.0–31.6), followed by Tijuana at 6.4% (95% CI 3.4–11.6), and Mexicali at 5.5% (95% CI 2.9–10.3). Overall IgG seroprevalence was 21.1% (95% CI 17.4–25.2), with Mexicali at 26.0% (95% CI 20.6–32.3), Ensenada at 21.9% (95% CI 16.2–28.9), and Tijuana at 18.7% (95% CI 13.7–25.0). Combined RT-PCR or IgG positivity was 26.3% (95% CI 22.2–30.9) overall. Border crossing in the past 6 months was not associated with SARS-CoV-2 prevalence (unadjusted OR 0.40, 95% CI 0.12–1.30). After adjustment, factors associated with higher IgG seropositivity included current smoking (adjusted pOR 3.0, 95% CI 1.5–6.1), age ≥60 years (adjusted pOR 1.8, 95% CI 1.0–3.2), and high school education versus elementary or less (adjusted pOR 2.2, 95% CI 1.2–4.1). Regarding prevention practices, 86% reported using face masks very frequently, 74% reported hand washing very frequently, but only 42% reported adopting quarantine or social isolation very frequently. Vaccine hesitancy was substantial: 30.4% (95% CI 24.4–37.1) reported being very unlikely to get vaccinated, and only 45% reported being very likely to get vaccinated if the vaccine were freely available. Among those hesitant, 41% worried about side effects, 10% did not think the vaccine worked, and 4% worried about the speed of vaccine development.
**Clinical Implications:** This study demonstrated that in February 2021, most of Baja California's population remained susceptible to SARS-CoV-2 infection, with only 21.1% showing evidence of prior infection through IgG antibodies. The high active infection rate (7.8% by RT-PCR) combined with low seroprevalence indicated ongoing community transmission as the national vaccination campaign began. The finding that border crossing was not associated with higher infection prevalence suggests that border restrictions implemented in March 2020 may not have been the primary driver of transmission dynamics. The high level of vaccine hesitancy (over 50% expressing concerns about side effects or effectiveness) represented a major public health challenge requiring targeted health literacy interventions. The low adoption of quarantine and home isolation (only 42% very frequently) highlighted the need for structural support, such as financial incentives, to enable compliance with isolation recommendations. These findings underscore the importance of repeated population-based serosurveys to capture the dynamic nature of epidemic spread within territories and to inform evidence-based public health policy, particularly in border regions with unique demographic and economic characteristics.