Ebola study notes afebrile patients, calls into question WHO criteria
Researchers found that the World Health Organization (WHO) Ebola case definition has a specificity of only 31.5%, and they noted that 9% of Ebola patients reported neither a fever nor any Ebola risk exposure, calling into question WHO norms, according to a large study yesterday in The Lancet Infectious Diseases.
Researchers from Britain and Sierra Leone analyzed data on 850 suspected and 724 lab-confirmed Ebola patients who presented to the holding unit of Connaught Hospital in Freetown from May 29 to Dec 8, 2014. Fever or history of fever (n=599, 83%), intense fatigue or weakness (495, 68%), vomiting or nausea (365, 50%), and diarrhea (294, 41%) were the most common presenting symptoms in suspected cases.
Based on data from these patients, the investigators found the sensitivity of the WHO case definition to be 79.7%, which means about 20% of true Ebola cases would be missed (false-negatives). They found the specificity of the case definition to be 31.5%, which means 68.5% of patients who would be selected for admission would not actually have Ebola virus disease (false-positives).
In a subgroup analysis, 15 (9%) of 161 lab-confirmed Ebola patients reported neither a history of fever nor a risk factor for Ebola exposure.
The team also found that including these conditions increased the accuracy of the case definition: intense fatigue, confusion, conjunctivitis, hiccups, diarrhea, or vomiting. A combination of three or more of any these symptoms tripled the odds of a confirmed Ebola diagnosis by yielding a sensitivity of 58% and specificity of 71%.
The authors conclude, "The finding that 9% of Ebola virus disease cases had no fever or history of fever and no risk factor for exposure to the virus shows that existing case definitions that typically include these features as criteria (eg, the existing WHO case definition) have insufficient sensitivity to identify all cases of Ebola virus disease." They add that existing clinical criteria cannot discriminate Ebola from other diseases like malaria and typhoid fever.
They write, "A highly specific Ebola virus disease screening algorithm could reduce the risk of patient-to-patient transmission within holding units, which is a major clinical concern."
A related commentary says previous studies have also found the WHO case definition to lack sensitivity and warns that overreliance on it can lead to increased healthcare spread.
Jul 23 Lancet Infect Dis study
Jul 23 Lancet Infect Dis commentary
US study finds 18% uneasy toward Ebola-volunteer colleague
Almost one in five public health colleagues of a volunteer physician who returned to New Hampshire from West Africa's outbreak region expressed discomfort with the responder returning to work even though she was considered low-risk, according to a study today in Emerging Infectious Diseases.
Public health colleagues in the infectious disease program, however, expressed no such unease.
New Hampshire researchers conducted a Web survey involving 178 New Hampshire Division of Public Health Services (DPHS) staff members (71.2% response rate) after the physician returned to the DPHS after her volunteer stint at Ebola treatment units in Sierra Leone. The survey included two scenarios: that the volunteer physician had contact with Ebola patients or did not have contact with Ebola patients.
Even given the no-contact scenario, 32 respondents (18%) said they would be uncomfortable with the responder returning to work. That rate rose to 35% and higher—even if the volunteer had no Ebola patient contact—in these situations: helping her if she fainted (35.0%), sitting next to her (35.6%), eating food made by her (37.6%), and attending a party at her home (46.7%).
If the physician had contact with Ebola patients, the percentage of coworkers expressing discomfort in those four situations rose to 53.7%, 49.7%, 54.0%, and 70.6%, respectively. And 42.7% of coworkers would be uncomfortable if she showed up for work after having contact with Ebola patients.
Responses, however, varied by professional background, with no worker in the DPHS infectious disease program showing discomfort with the volunteer returning to work, despite the fact that this group would have the most contact with her. Those not in the infectious disease program had an almost 11 times higher likelihood of expressing discomfort, and those with an education below a bachelor's degree had an almost three times higher likelihood.
Jul 24 Emerg Infect Dis study