Delayed first contact in a universally insured population: time to care and COVID-19 outcomes in Mexican primary care

Time from symptom onset to first care may be a modifiable determinant of COVID-19 outcomes, but it is rarely quantified in primary care. This retrospective cohort study included all 5,248 confirmed SARS-CoV-2 cases (RT-PCR or certified rapid antigen test) notified to the national surveillance system (SINOLAVE) by a large family medicine unit of the Mexican Social Security Institute (IMSS) in Leon, Mexico, between April 2020 and April 2021. Pandemic waves were classified using the official IMSS definition. Wave profiles were compared with chi-square tests and Bonferroni correction, and predictors of death were assessed by multivariable logistic regression, with sensitivity analyses adjusting for severity signs at first contact. Of the cases, 53.8% were female, mean age was 41.1 years, and 317 deaths (6.0%) were registered. Median time to first care was 3 days (interquartile range: 2-5). Case fatality was 3.0% among patients presenting within 3 days and 9.1% among those presenting later. Later presentation (> 3 days) was associated with death (adjusted odds ratio = 1.98; 95% confidence interval: 1.47-2.65), but the association was attenuated after adjustment for severity signs at first contact (adjusted odds ratio = 1.20; 95% confidence interval: 0.85-1.69), which were more frequent among later presenters (21.8% vs. 9.3%). From the first to the second wave, fever (76.1% to 56.9%), diabetes (22.8% to 9.6%), and case fatality (10.5% to 4.8%) decreased, whereas anosmia and dysgeusia increased (all p < 0.001). Surveillance data cannot distinguish clinical progression during delay from late presentation of already severe patients. Delay persisted despite universal entitlement to care, indicating that insurance coverage alone did not ensure timely presentation; prospective cohorts measuring severity at onset are needed.