An evaluation of the Zambia influenza sentinel surveillance system, 2011–2017

Simusika, Paul; Tempia, Stefano; Chentulo, Edward; Polansky, Lauren; Mazaba, Mazyanga Lucy; Ndumba, Idah; Mbewe, Quinn K.; Monze, Mwaka

• Burden of influenza-associated illness (Yes) [14] Sustainability Each quantitative indicator was evaluated as the proportion (expressed as percentage) of the outcome of interest over the total. A scale from 1 to 3 was used to provide a score for each quantitative indicator as follows: < 60% (from the above calculation) scored 1 (weak performance); 60-79% scored 2 (moderate performance); ≥80% scored 3 (good performance). For indicators for which a proportion over a total could not be obtained (qualitative indicators) a score was assigned based on the same scale using expert consensus b Indicator value calculated by dividing the number of achieved outcome by the total number of outcome considered (i.e. 4/5 = 80.0% or 1/4 = 25.0%) influenza-associated illness across different geographical areas of the continent may be expected [15] . Increasing the number of surveillance sites in Zambia should be considered in light of increased cost, logistical constraints (e.g. regular supervision and transportation of specimens) and impact on laboratory testing capacity and potentially on data quality. Instead of expanding the geographic coverage of the system, efforts could be targeted to improve timeliness of sample shipment and enrollment of patients meeting the SARI case definition at the existing sentinel sites as well as collecting more detailed epidemiological information on enrolled cases (e.g., relevant risk factors for influenza-associated severe illness and in-hospital outcomes that are recommended by the WHO guidelines for global influenza surveillance, but are currently not fully captured by the system). In particular, the collection of additional information on underlying medical conditions may enable the continuous monitoring of high risk groups for influenzaassociated severe illness. This would also align the Zambia ISSS to the minimum data collection requirements suggested by WHO [6] . The flexibility of the system was demonstrated by its ability to investigate the viral etiology of pathogens other than influenza [24] . While the recruitment of dedicated surveillance officers may not be justifiable for a vertical influenza surveillance program it may be considered for an integrated multi-pathogen surveillance platform. This could be potentially cost-beneficial for a sentinel surveillance system where a limited number of surveillance officers may be needed in selected sentinel sites as compared to national passive surveillance programs. Integrated and well-run multi-pathogen surveillance systems, if adequately staffed may provide high quality and timely epidemiological and virological data with a "modest" financial investment. The national burden of illness attributable to specific pathogens can be estimated from sentinel surveillance data through special studies as demonstrated in Zambia and other African countries for influenza [10] [11] [12] [13] [14] [15] [16] . This would be particularly relevant for policy makers if multiple pathogens are included so as to provide the relative burden associated with different etiological agents. Risk factors for influenzaassociated severe illness (e.g. hospitalization or death) have also been assessed using sentinel surveillance systems [23, [25] [26] [27] and this could be implemented also in Zambia for influenza and other pathogens.