GENEVA (AFP) - The raging Ebola outbreak has likely killed far more people than the 4,818 deaths reported by the World Health Organisation (WHO), an expert at the UN health agency said Thursday, warning that thousands of fatalities were likely not accounted for.
"There are lots of missing deaths in this epidemic," Christopher Dye, WHO's strategy chief, told AFP, estimating that around 5,000 fatalities could be missing from the count.
This assessment, he said, was based on the knowledge that the fatality rate in the epidemic centred in Guinea, Liberia and Sierra Leone stands at about 70 per cent.
But with total reported cases of infections reaching 13,042, that suggests that many of the deaths were going unrecorded.
Dye said the likely explanation was that many people were burying the dead in secret, possibly to avoid having authorities interfere with burial customs like washing and touching the deceased widely blamed for much of the transmission.
The UN's health agency has created confusion with its latest figures of Ebola cases and deaths, which have shown shrinking numbers.
The toll provided Wednesday night showed 4,818 deaths, down from 4,951 reported on Oct 31, while the number of reported cases fell to 13,042 from 13,567.
This does not mean that the epidemic is over or that people have stopped dying from the deadly virus, Dye said, explaining instead that the drop in numbers was linked to a shift in the way WHO uses different databases to calculate the overall numbers.
"Many, many people are still dying of Ebola," he said.
Up until recently, WHO had used several different databases from each of the affected countries to calculate the overall number of cases and deaths.
The different sources however meant that the numbers did not always progress in a consistent manner, fluctuating according to what data was available from the separate databases.
To avoid the fluctuations, the agency had shifted to only using data from each of the countries' situation reports, based on daily counts of patients and deaths district by district.
Dye said the new method, which entailed the significant drop in deaths in WHO's last overview, was "slightly less satisfactory" in terms of accurately reflecting the numbers of cases and deaths, but was "nonetheless always consistent."
He stressed though that the cumulative total was far less interesting than graphs WHO provides in its updates showing the progression of the epidemic week by week in each of the three hardest-hit countries.
Those graphs, which are still based on several datasets, show clearly that "there are downward trends in some parts of the epidemic area," Dye said, adding "I think that's beyond question now."
WHO confirmed last week that Liberia, the country hardest-hit by the outbreak with more than 6,500 cases and nearly 2,700 deaths, was showing a slowdown in transmission.
Dye said that trend appeared to be continuing across the country, while numbers in much of Guinea looked "somewhat flat".
Numbers also seemed to be flattening out in some parts of Sierra Leone, although there were still "lots of worries" about transmission in the west of that country, he said The trend is good news, since it means the Ebola outbreak is not on the worst-case-scenario trajectory used for shocking forecasts a few months ago by WHO the US Centres for Disease Control (CDC), Dye said.
"There was talk of millions of cases" by January, he pointed out.
"That's just because if you have an upward curve and you keep on projecting it you're going to get millions of cases," he said, stressing "it is pretty clear to us now that that is not going to happen."