Should mortality data for the elderly be collected routinely in emergencies? The practical challenges of age-disaggregated surveillance systems

Data on the elderly are rarely collected in humanitarian emergencies. During a refugee crisis in South Sudan, Médecins Sans Frontières developed a prospective mortality surveillance system collecting data for those aged ≥50 years and found that the elderly were dying at five times the rate of those aged 5-49 years. Practical and ethical issues arose. Were reported ages accurate? Since no baseline exists, what does the mortality rate mean? Should programmatic changes be made without evidence that these would reduce the elderly mortality rate? We outline issues to be addressed to enable informed decisions on response to elderly populations in emergency settings.