Public Release: 

New model for early meningitis detection


A meningitis incidence threshold of 10 cases per 100 000 inhabitants in just 1 week - lower than previously recommended by the WHO - can reliably be used to confirm an epidemic in time to implement vaccination programmes, conclude authors of a study in this week's issue of THE LANCET.

Epidemics of meningitis in sub-Saharan Africa cause serious illness in tens of thousands of people, and thousands of deaths during epidemic years. The disease is often detected too late for the implementation of successful vaccination programmes. Rosamund Lewis and colleagues from Epicentre, Paris, France, assessed weekly meningitis incidence as a tool to detect epidemics in time to implement mass vaccination.

The incidence of meningitis for 41 subdistricts in Mali, west Africa, was determined from cases recorded in health centres between 1989 and 1998, and from surveillance data between 1996 and 1998. For incidence thresholds of 5 to 20 cases per 100 000 inhabitants per week, the investigators calculated sensitivity and specificity for detecting epidemics, and determined the time lapse between threshold and epidemic peak.

9084 meningitis cases were recorded. Clinic-based weekly incidence of 5 and 10 cases per 100 000 inhabitants detected all meningitis epidemics (sensitivity 100%) with an average threshold-to-peak time of 5 and 3 weeks, respectively. Under-reporting reduced sensitivity: only surveillance thresholds of 5 or 7 cases per 100 000 inhabitants per week detected all epidemics. At 10 cases per 100 000 inhabitants per week, false alarms occurred on fewer than 2% of occasions that the threshold was crossed. For populations under 30 000, 3 to 5 cases of meningitis in one or two weeks accurately predicted subsequent epidemics.

Rosamund Lewis comments: "Low thresholds improve the timely detection of devastating meningitis epidemics in Africa. An alert threshold of 5 cases per 100 000 inhabitants per week allows time to investigate the outbreak, prepare for an epidemic, and initiate mass vaccination where appropriate. For areas with fewer than 30 000 people, two cases in a week should raise the alarm. High quality epidemiological surveillance is therefore essential to detect epidemics in time to respond and protect the population."

In an accompanying Commentary (p 255), Bradley Perkins from the Centres for Disease Control and Prevention, USA, states that future success in meningitis control may lie in new vaccines which can be given routinely in early childhood. He comments: "Until meningococcal conjugate vaccines containing serogroup-A are available, countries and partners (eg, non-governmental organisations, donors) should vigorously support the implementation of the revised WHO guidelines for detecting meningococcal meningitis epidemics in highly-endemic African countries. Efforts such as those of Lewis and colleagues are examples of high-quality applied public-health research critical for the evaluation and adaptation of practice to optimise the impact of interventions."


NB. Please note that if you are outside North America, the embargo for LANCET press material is 0001 hours UK Time Friday 27th July.

Contact : Dr Rosamund Lewis, WHO, PO Box 24578,Kampala, Uganda T) 256-75-721956; F) 256-41-344059; E) Dr Bradley A Perkins, Division of Bacterial and Mycotic Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, QGeorgia 30333, USA; T) 1-404-639-2215; F) 1-404-639-3970; E) BPerkins@CDC.GOV

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.