Elsevier

The Lancet

Volume 374, Issue 9707, 19 December 2009–1 January 2010, Pages 2072-2079
The Lancet

Articles
Infection and death from influenza A H1N1 virus in Mexico: a retrospective analysis

https://doi.org/10.1016/S0140-6736(09)61638-XGet rights and content

Summary

Background

In April, 2009, the first cases of influenza A H1N1 were registered in Mexico and associated with an unexpected number of deaths. We report the timing and spread of H1N1 in cases, and explore protective and risk factors for infection, severe disease, and death.

Methods

We analysed information gathered by the influenza surveillance system from April 28 to July 31, 2009, for patients with influenza-like illness who attended clinics that were part of the Mexican Institute for Social Security network. We calculated odds ratios (ORs) to compare risks of testing positive for H1N1 in those with influenza-like illness at clinic visits, the risk of admission for laboratory-confirmed cases of H1N1, and of death for inpatients according to demographic characteristics, clinical symptoms, seasonal influenza vaccine status, and elapsed time from symptom onset to admission.

Findings

By July 31, 63 479 cases of influenza-like illness were reported; 6945 (11%) cases of H1N1 were confirmed, 6407 (92%) were outpatients, 475 (7%) were admitted and survived, and 63 (<1%) died. Those aged 10–39 years were most affected (3922 [56%]). Mortality rates showed a J-shaped curve, with greatest risk in those aged 70 years and older (10·3%). Risk of infection was lowered in those who had been vaccinated for seasonal influenza (OR 0·65 [95% CI 0·55–0·77]). Delayed admission (1·19 [1·11–1·28] per day) and presence of chronic diseases (6·1 [2·37–15·99]) were associated with increased risk of dying.

Interpretation

Risk communication and hospital preparedness are key factors to reduce mortality from H1N1 infection. Protective effects of seasonal influenza vaccination for the virus need to be investigated.

Funding

None.

Introduction

At the beginning of April, 2009, the medical care units of the Mexican Institute for Social Security (Instituto Mexicano del Seguro Social, IMSS) were alerted because the number of seasonal influenza cases did not decrease as expected from March to May. Additionally, three outbreaks of influenza-like illness were reported in the Mexican States of Veracruz, Tlaxcala, and San Luis Potosí from March to April.1 These events, in conjunction with a report2 of a suspected case of non-typical pneumonia in the State of Oaxaca on April 15, triggered both an epidemiological alert on April 17, and intensified surveillance of severe acute respiratory infections in inpatients. On April 23, Mexican officials announced that a novel influenza A H1N1 virus (pandemic H1N1) had been identified in two samples—one from the outbreak in Veracruz3 and another from Oaxaca. By then, 18 confirmed H1N1 cases had been reported.4

By Sept 27, more than 4100 deaths were associated with the pandemic worldwide,5 with 3020 deaths in the Americas mainly occurring in the USA, Argentina, Mexico, Brazil, and Canada. Mexico has one of the highest numbers of registered deaths at that date for this pandemic, with 146 cases6—63 of which have been reported by IMSS. IMSS is a Mexican public institution that provides health-care services to a population of nearly 40 million people, manages 1099 primary health-care units and 259 hospitals across the country, and has reported the largest number of cases and deaths of H1N1 within the country. In a pandemic, infection and death rates are expected to affect countries in different ways.7 Individual host factors (eg, immune function, nutritional status, acquired immunity through previous influenza infection, and comorbidity) and community factors (population density and mixing rates, quality and access to health care, and the physical environment) can explain variations in mortality between communities.7

The epidemic in Mexico had a transmissibility rate of 1·2% and an estimated disease-specific mortality rate of 0·4%;8 however, in view of new data, ranges for mortality have been set between 0·20% and 1·23%, with the lowest rates reported in the European Union (EU) and the highest in Mexico. At present, the pandemic has spread to more than 168 countries.5 We therefore need to stay alert—especially in countries with similar sociodemographic characteristics to Mexico, which might share conditions that could potentially contribute to H1N1 mortality. An important step is to assess international strategies and programmes to control domestic epidemics, such as the controversial school-closure policy to prevent transmission.9, 10 Mexico has been the only country to shut down the school system nationwide—from nurseries to universities—at the beginning of the pandemic.

We report the IMSS experience of the timing and spread of H1N1, and investigate some protective and risk factors for infection, severe disease, and death.

Section snippets

Data collection and surveillance

Before the H1N1 epidemic, Mexico had an active but incomplete influenza surveillance system. This surveillance allowed detection of the first cases of the new influenza but was not effective in assessment of the extent of the epidemic. After the first alert on April 17, active surveillance for severe pneumonia was started in all IMSS hospitals. Therefore, surveillance was only of patients in hospital, although almost all reported cases up to April 28, were the most severe.

On April 28, IMSS

Results

The first large outbreak of H1N1 in Mexico affected the Mexico City metropolitan area, San Luis Potosí, and Zacatecas in April, 2009, and lasted until June 6, 2009. Figure 2 shows results of intensive surveillance for influenza-like illness after April 28, in which up to 2800 cases were reported on May 2, during the peak of the epidemic. Since laboratory capabilities were initially overwhelmed, the number of confirmed cases as a proportion of suspected cases was very low. During this first

Discussion

Our surveillance analysis showed a large outbreak in April, 2009, that mainly affected the population in Mexico city. A second large outbreak took place in the southeast of Mexico during June and July. Mortality rates from pandemic H1N1 differed from those reported in other countries, probably because Mexico was the first country to have intensive virus circulation. The disease-specific mortality rate was high during the first outbreak but was similar to that recorded in other countries in the

References (25)

  • Pandemic (H1N1) 2009—update 60

  • Situación actual de la epidemia de influenza A (H1 N1)- Comunicado de prensa no 252. 2009. [in Spanish]

  • Cited by (0)

    View full text