There is increasing interest in using real-world data (RWD) in public health and epidemiology. This study presents a methodological exercise applying three observational study designs—cross-sectional (CS), case-control (C-C), and cohort—to the same dataset to explore their practical application in evaluating the association between COVID-19 vaccination and all-cause mortality. A case study was conducted using data from the Respiratory Disease Epidemiological Surveillance System in Mexicali, Baja California, Mexico, from March 2020 to February 2022. Confirmed COVID-19 cases were included; death was the outcome, and vaccination status the exposure. CS analysis applied Poisson regression to estimate prevalence ratios (PR). C-C analysis selected deaths as cases and 1:2 matched controls, using logistic regression for odds ratios (OR). Cohort analysis used symptom onset as entry date, with Cox regression estimating incidence rate ratios (IRR). Analyses were performed with R 4.2.3. Of 47,340 cases, 71.3% were unvaccinated, 2.07% had an incomplete schedule, and 26.6% had a complete schedule. CS and cohort analyses showed PR and IRR of 0.48 (95%CI: 0.36–0.63) for incomplete schedule, and 0.29 (95%CI: 0.26–0.32) for complete schedule. C-C analysis showed OR of 0.29 (95%CI: 0.19–0.43) and 0.15 (95%CI: 0.13–0.18), respectively. RWD enables replication of multiple observational designs; however, its use requires strategies to address data heterogeneity, potential bias, and quality. When handled with proper methodological care, RWD can support timely and context-relevant public health decision-making.
Ovalle-Marroquín et al. (Wed,) studied this question.