It is an uncomfortable fact. Only a little more than one-in-eight Samoan households’ drinking water has been found to contain as much E.coli bacteria as water.

This is, in our view, a national problem of major significance and one that deserves immediate action. 

These figures were revealed in stories carried on the front page of the Observer last week, most recently on Wednesday. (“Stats chief clarifies water contamination; experts call for investigation”.) 

There are subtleties to this story. Subtleties that we intend to acknowledge here but which should not let us lose sight of the need for action. 

E.coli is a proxy for contamination of water, most often with faecal matter. 

When we first published a story that 47 per cent of Samoan households had E.coli contamination in their water, following the release of a report jointly compiled by our national statistics bureau; The United Nations Children's Fund; the U.N. Population Fund and the Pacific Community, the figures seemed shocking.

Experts, such as Australia’s foremost expert in drinking water safety Professor Stuart Khan from the University of New South Wales in Australia, told us that a level of one per cent might be expected to be found in a survey.

“[Almost] fifty per cent tells you that there’s something really wrong,” Dr. Khan said. We knew this was a story deserving of national attention.

But as we made clear in that initial report carried in Monday last week’s Observer, there was no indication of how much contamination there was. 

When we finally received a response from the Government we realised that the problem was much more serious than we had ever first imagined.

The Government’s Water Authority had not answered our questions in the lead up to the publication of the first story.

When the Government Statistician wrote a letter to our paper they provided much greater insight into the depth of the problem in our households. 

Eliminated was the possibility that these households had only registered low levels of contamination in their drinking water. Instead nearly 18 per cent were found to be moderately contaminated with E.coli (up to 10 mls for every 100); 15.4 per cent with counts classified as high (between 11-100 mls) and then the aforementioned 13.7 per cent very high (100 ml per 100 ml of water). 

The facts are simple: there are no safe levels of E.coli contamination in any human drinking water and the guidelines for countries such as New Zealand make this clear. No mililitres per one hundred militres. 

E.coli contamination. is one of the most reliable measurements of water being polluted with faeces of some kind and so a test often performed by scientists as a measure for the health of a nation’s water supply.

But constant, long-term exposure to E.coli itself - while it may not leave those who drink it with the sour stomachs that it does to those who have not built up a resistance to it - have been found to have higher levels of high blood pressure and kidney problems.

We should take a moment here to acknowledge the Samoa Bureau of Statistics’ response to our story and some of the other objections raised. 

The story was based on the release of a preliminary stage of the Samoa Demographic and Health Survey-Multiple Indicator Cluster Survey 2019-20.

As part of that process they measured the drinking water of 696 households who had been randomly selected. 

“The number of sampled households which have been interviewed to test the drinking water quality was only 696 households,” the Government Statistician, Ali’imuamua Malaefono Ta’aloga, said in a letter to “clarify our story”.

“This is about 2% of 28,880 households at the time of the survey.”

We accept and always reported as such that this research was based on a survey of households. Not a census.

But it was a survey designed with technical support from two different agencies of the United Nations. 

The science of statistics tells us that so long as you are taking your samples from the right places, these can represent a much bigger number of households.

In the realm of politics, for example, opinion polls of 328 million American voters are often based on samples of as few as 1,000 voters, so long as they are chosen to be representative.

For the survey on which the story was based, not only was technical support provided by two United Nations, the sample certainly appears to have been based on sound sampling according to the report in question: 

“A systematic sample of 20 households from rural [areas] and 15 households from urban [areas] was drawn,” the report said.

When asked if she considered this sample to be statistically valid, Ali’imuamua said she would respond in due course. 

We suspect that it is. And even being conservative and accounting for the possibility of a significant margin of error, these statistics do not lose their power to shock. 

This is not to point fingers but rather to hopefully to promote action in the name of helping those who need it most. 

The Samoa Water Authority said that it is seeking to make greater investments in treating parts of the country that do not have access to treated water. We applaud them for that. 

Also the entire nation does not receive its water from the S.W.A. Up to 12 per cent do so from independent water schemes. And, as experts in our story were quoted as saying, a significant amount of contamination appears to be occurring as water makes its way from the source to the household tap. 

Something needs to be done; that is the point of this story. A good place to start might be a review of all of S.W.A.'s non-critical expenditure including its sponsorship of $200,000 for weightlifting.

But more broadly, the Government might take this as an occasion to review its priorities. 

So long as a significant amount of our citizens are going without clean access to water, the giver of life, how much pride can our Government take in the next unveiling of a piece of infrastructure?