Laboratory-based surveillance in Latin America: attributes and limitations in evaluation of pneumococcal vaccine impact

Disease surveillance data are needed to monitor trends in disease activity, inform decision-making in public health and evaluate disease prevention/control measures. The Sistema Regional de Vacunas (SIREVA) supports laboratory-based surveillance of invasive pneumococcal disease (IPD) in Latin American countries, providing information on identification, distribution, and anti-microbial susceptibility of pneumococcal strains. We estimated the proportion of pneumococcal meningitis and sepsis/bacteremia cases captured by SIREVA, by comparing the number of SIREVA-reported isolates in Argentina, Brazil, Chile, Colombia, Ecuador and Mexico with the estimated expected number of cases based on regional estimates of disease incidence. In all six countries, the number of isolates reported by SIREVA was consistently lower than the number of cases expected, across all years with data available. The proportion of SIREVA-reported isolates was highest in Chile (43-83%) and lowest in Mexico (1.4-3.5%). Passive surveillance systems such as SIREVA are important tools for monitoring circulating strains that could be related to pneumococcal disease, but our results show that SIREVA is likely to underestimate pneumococcal disease incidence. This under-reporting will limit the precision of surveillance data in monitoring changes in the incidence of IPD after vaccine introduction, and this should be considered when assessing the impact of vaccination programs. PLAIN LANGUAGE SUMMARY What is the context? Infectious disease surveillance is an important epidemiological tool to monitor the health of a population. Surveillence can be used to detect trends in disease activity and to trigger disease control measures. In Latin America, the SIREVA surveillance system monitors occurrence of bacterial pneumonia, sepsis/bacteremia and meningitis. However, passive surveillence systems may understimate disease occurrence. What is new? We compared the number of isolates of invasive pneumococcal disease (IPD), specifically meningitis and sepsis/bacteremia, in children aged <5 years reported in SIREVA data in six countries in Latin America with the expected number of cases based on regional estimates of IPD incidence. Our results show that the number of isolates reported by SIREVA was consistently lower than the estimated number of cases, across all six countries and all the years available. The percent difference between SIREVA-reported isolates and estimated number of cases was variable between countries, ranging from 43-83% in chile to 1.4-3.5% in Mexico. What is the impact? Passive surveillance systems such as SIREVA are important tools for monitoring disease incidence, but they are likely to underestimate pneumococcal disease occrruence. This under-reporting will limit the precision of surveillance data in monitoring changes in disease incidence after vaccine introduction, and this needs to be considered when assessing vaccine impact.