Comparative epidemiology of five waves of COVID-19 in Mexico, March 2020–August 2022

SARS-CoV-2 has spread worldwide in the form of a series of waves of infection with varying levels of transmissibility and severity [1, 2] and modulated by multiple factors, including the intrinsic transmission properties of evolving variants, social distancing measures (e.g., face mask wearing), access to medical care and vaccination campaigns. Investigating the epidemiological factors that have shaped the multi-wave COVID-19 patterns of infection, hospitalization, and death at different spatial scales could improve public health policy to save lives and minimize economic disruptions for future pandemics.

The COVID-19 pandemic has disproportionately affected many Latin American countries, with 176,521,913 cases, including 2,822,535 deaths reported as of September 8th, 2022 [3]. Mexico, one of North America’s most highly populated countries with high poverty levels [4] and low COVID-19 testing rates, documented the first imported COVID-19 case on February 27th, 2020. Mexico currently ranks 7th in the world regarding COVID-19 reported deaths, with a total of 329,652 recorded deaths and 7,046,220 reported cases (4.68% of confirmed case fatality rate) as of September 8th, 2022 [5].

In this study, we investigated the epidemiological patterns of COVID-19 infections, hospitalizations, deaths, and factors associated with severe disease outcomes during the five epidemic waves in Mexico. To that end, we used data reported by the Mexican Institute of Social Security (IMSS) surveillance system, the most extensive Latin-American social security system, and Mexico’s leading health institution [6].

IMSS confirmed the first case at the end of March 2020. The epidemiological system reported symptomatic and laboratory-confirmed cases. The number of cases and estimates of hospitalization and in-hospital case fatality rates are presented in Table 1.

The five waves comprised 8,783,565 suspected cases, of which 7,502,195 were tested (85.4%) and 3,396,375 laboratory-confirmed COVID-19 cases with onset of symptoms from March 29th, 2020, to August 27th, 2022 were analyzed (positivity rate of 45.3% among tested cases). Overall, 11% cases were hospitalized (373,646 of 3,396,375 confirmed cases), 20.6% were intubated (76,929 of 373,646 hospitalized cases), and 45.1% was the overall hospital case fatality rate (168,499 of 373,646 hospitalized cases). Below we describe the epidemiological patterns related to case incidence rates of confirmed cases, hospitalizations, in-hospital mortality, hospital admission delays, and in-hospital stays across waves, age, and gender.

The COVID-19 pandemic has been characterized by a multi-wave pattern driven by multiple factors, including social control measurements, SARS-CoV-2 variants, testing practices, vaccination, and health care infrastructure. In this report, we described the epidemiological patterns of COVID-19 during five epidemic waves from surveillance data reported by IMSS, the most extensive Mexican and Latin-American social security system. Roughly, we observed an increase in reported incidence and a decrease in the severity of COVID-19 confirmed cases over time. Additionally, based on a subset of samples, we found that the second, third, fourth, and fifth waves were dominated by the Alpha-Gamma, Delta, and Omicron SARS-CoV-2 variants, respectively. During the first three waves, Mexico experienced a high in-hospital mortality rate associated with hospitalization criteria for critical patients with comorbidities. Further research is needed to improve case management during future pandemics.

Although surveillance systems do not represent all disease cases, they provide time patterns and valuable information for policy decision-making. Confirmed cases represent only a sample of all cases because not all ill persons seek medical care and have a specimen collected [10]. Additionally, the test’s sensitivity and specificity depend on the collection timing and the specimen’s quality. This under-ascertainment of COVID-19 ill persons has been evaluated with seroprevalence and excess mortality studies [11‐13]. A seroprevalence survey just before vaccination began in December 2020 estimated the extent of SARS-CoV-2 infections in Mexico, regardless of disease severity and test availability, at as much as 25%, from a national representative sample [11]. The excess mortality percentage in Mexico has been estimated at 32% across the 2 pandemic years, and the number of death certificates associated with COVID-19 has been ~ 1.5 times the number of laboratory-confirmed deaths reported in the epidemiologic surveillance system [12].

In this case, trends of positive cases are representative of testing recommendations, adequately represent critically ill patients, and underrepresent ambulatory symptomatic cases. Although the recommendations for testing were similar during the five waves, the introduction of rapid testing presented a differential practice after the second wave (Additional file 1: Figure S7). Therefore, the magnitude of the first and second waves could be underestimated, and the comparison with subsequent waves may be affected by under ascertainment bias.

During the first wave in Mexico, the hospitalization percentage among 264,792 confirmed cases was 39.9% and declined to 1.8% in the fifth wave. The first figure was similar to the hospitalization percentage of 31.9% reported in March–August 2020 in the United States [14]. We also observed a higher hospitalization percentage in males and older people. In this respect, the Palmer et al. study reported that the exponential increase in COVID-19 hospitalization rates with age could be associated inversely with thymic T-cell production, which declines in males and older age [15]. In line with our results, a Denmark study observed a reduction in hospitalization rates for Omicron cases compared with Delta variant cases [16]. Our higher hospitalization rates compared with this previous study could also be explained by the Mexican testing strategy, which has been focused on severe COVID-19 cases.

Our in-hospital case fatality rate estimates were higher than other reports worldwide [17] but similar to those reported for critically ill hospitalized patients [18, 19]. A combination of factors could explain these elevated in-hospital case fatality rate estimates. These include the hospitalization practices characterized by strict admission criteria, admitting mostly critically ill cases, healthcare resources-related factors, the high vulnerability of the sick population, and the low COVID-19 vaccination coverage during the first waves (Additional file 1: Figures S6, S8, and S9). The contribution of organizational factors, the low proportion of intubated patients during hospitalization, and the reduced number of patients admitted to ICU deserve further research.

The multivariate logistic model results confirm the independent contribution of gender, age, previous medical conditions, vaccination status, and the reduction of disease severity over time. Similar to results from Milan, Italy, reported by Giacomelli et al., the hospital case fatality rate had its highest value during the second wave [20]. De la Peña reported that the case fatality rate increased with diabetes, hypertension, obesity, immunosuppression, and end-stage kidney disease. The population attributable fraction due to obesity in outpatients was 16.8% [21]. In another study, Hong et al. [22] reported that Mexico was among the countries with the highest CFRs. They also mentioned that medical resources might be important in preventing COVID-19-related deaths; however, considering the slight variation in fatality among the elderly, preventive measures such as vaccination are more critical, especially for the elderly population, to minimize the mortality rates. Olivas-Martinez et al. [23] found that the mortality rate over time was related to the availability of ICU beds, indirectly suggesting that overcrowding contributed to hospital mortality. In contrast with other studies that report an increase in COVID-19 severity with delta strain [24, 25], our study found that the third wave, mainly driven by this variant, had a lower in-hospital case fatality rate than the first two waves. These differences could be explained by the abovementioned hospitalization criteria and testing practices during the first two waves and the COVID-19 vaccination strategy that initially targeted the most vulnerable groups, leading to a decline in hospitalization rates for high-risk groups.

As expected, we found an important contribution of vaccination in preventing severe disease and reducing the risk of hospitalization. The OR for hospitalization among confirmed cases was 0.41 (95% CI 0.40, 0.41) compared to unvaccinated or incomplete vaccination. The impact of vaccination on the population level requires additional information for the general population. This information was not available in this study, restricted to laboratory-confirmed cases. However, the impact of vaccination on the COVID-19 pandemic has recently been reported with promising results [26‐28].

A strength of our study is the large sample size, including records for nearly 3.4 million laboratory-confirmed COVID-19 cases during a 2-year pandemic period stemming from a population of 60 million individuals, representing the Mexican Population. Nevertheless, our findings need to be contextualized, considering some limitations. The analyzed data derives from passive database systems, which collect information from patients that seek and receive medical care at IMSS health care facilities. In this respect, milder COVID-19 cases would not be fully represented. Additionally, specific information about administering specific therapeutics before or during hospitalization was unavailable. With our data, we could not distinguish between hospitalization due to COVID-19 and hospitalizations for reasons other than COVID-19. Moreover, only vaccination coverage at the Mexican population level was available, with no information on specific coverage for IMSS affiliates.

In summary, our findings indicate that reported incidence tended to increase during the five COVID-19 epidemic waves experienced by the Mexican population. Still, severity tended to decline over subsequent waves, with an important effect of vaccination. The high in-hospital mortality rate was associated with hospitalization criteria for critical patients with comorbidities during the first three waves.