Thursday, August 25, 2016

EID Journal: Estimation of Severe MERS Cases in the Middle East, 2012–2016

Credit CDC


One of the realities of disease surveillance is that the number of cases `officially' reported by public health agencies is assumed to represent only a fraction of the total number of cases in a populaton.

Factors affecting undercounts can include the percentage of mild cases (unlikely to be tested), the availability and quality of local medical care, the sensitivity of lab tests, the amount of background noise from similar diseases, and whether public health is actively `looking' for a specific disease. 

It is probably fair to say a novel flu infection is far more likely to be detected (and reported) by Hong Kong authorities than by those from rural Mainland China, Cambodia, or Central Africa.

We've looked at these challenges before, and while they may be more pronounced in low resource regions of the world, even our own CDC can't tell us how many people get the flu, or Lyme Disease, or West Nile Fever each year in the United States. 

Trying to get a handle on the burden and spread of emerging infectious diseases - particularly those with some pandemic potential - is always a priority, and in recent years the MERS Coronavirus has been very near the top of that list.

We've previously looked at estimates that have suggested the number of MERS cases far exceeds the numbers reported by countries on the Arabian Peninsula.   

The expectation is that many of these `uncounted' cases were mild, and were therefore never hospitalized or tested. Complicating matters, the persistence of antibodies in `mild' MERS-CoV infections appears limited, so the sensitivity of seroprevalence studies may be limited as well.

We've a new study, published today in the EID Journal, that makes another attempt to estimate the number of severe MERS-CoV cases in the Middle East since 2012, based on a small number of infected travelers from the region.

The authors estimate 3,250 (95% CI 1,300–6,600) severe MERS cases have occurred since 2012 on the Arabian peninsula.  

In other words, their estimate of severe cases is 2.3-times higher than the total number of all laboratory-confirmed cases combined (including mild and asymptomatic) cases in the region.

Granted, their results are based on a small (n=11) number of infected travelers, extrapolated out across the population of the region. Still, their multipliers are actually fairly low, compared to others we've seen for MERS and novel influenzas.

I've included the abstract and excerpts from the discussion, but follow the link below to read the report in its entirety.

Estimation of Severe Middle East Respiratory Syndrome Cases in the Middle East, 2012–2016 

Justin J. O’Hagan Comments to Author , Cristina Carias, Jessica M. Rudd, Huong T. Pham, Yonat Haber, Nicki Pesik, Martin S. Cetron, Manoj Gambhir, Susan I. Gerber, and David L. Swerdlow


Most Middle East respiratory syndrome cases have been recorded in the Middle East. Using data from travelers to this region, we estimated 3,250 (95% CI 1,300–6,600) severe MERS cases occurred in the Middle East during September 2012–January 2016, which is 2.3-fold higher than the number of laboratory-confirmed cases recorded in these countries.

Middle East respiratory syndrome (MERS), caused by MERS coronavirus (MERS-CoV), was first recognized in September 2012 (1). From that time until January 2016, >1,600 cases were laboratory-confirmed, and ≈600 deaths have been attributed to the virus (2). Cases have been detected among persons who traveled from the Middle East to 16 countries, and a MERS-CoV outbreak in South Korea introduced by a traveler caused >100 cases (3).

Estimates of the epidemic size in the Middle East are required to understand the level of MERS-CoV circulation and the likelihood of MERS-CoV exportations. However, these estimates have not been calculated for >2 years, during which time the number of recorded cases has increased by >15 times (2,4). We used data from travelers to this region to update estimates of severe MERS cases in the Middle East.

Our estimates were based on a small sample size (11 travel-associated cases) and assumed that travelers and residents of the Middle East had similar infection risks. Our sensitivity analyses demonstrated that results are sensitive to travelers’ estimated lengths of stay and also showed that estimates of the epidemic size that incorporated data from lower-income countries were 60% lower than estimates obtained by using data from high-income countries alone. This finding implies different levels of case detection across travelers’ home countries or different MERS-CoV exposure between visitors of different nationalities. Additional data (e.g., larger sample size; travel volume; and lengths of stay stratified by age and immigration status, frequencies of testing and contact with camels) could provide further estimates.

Public health officials are concerned about MERS-CoV, both in the source countries and from exported cases in persons who can seed outbreaks elsewhere (9,11). By better estimating the epidemic size in the Middle East, our results can help guide public health preparedness efforts in source countries and contribute to projections of the number of cases that could occur among travelers (9,11–13).
Dr. O’Hagan is an epidemiologist and transmission modeler at Centers for Disease Control and Prevention. His interests include the combined use of mathematical modeling and epidemiologic studies to better understand the effect of infectious disease risk factors and interventions.