Editors:

Further to your article "Tests reveal tap water 'drinkable'" (Nov. 19-Dec.

2) by Maeve Donelan.

We are afraid that rather than providing new and useful information to your readers,

your article can contribute to unjustified complacency or alarm. It also contained

serious errors and a number of omissions.

Water that is bacteriologically "safe" is not necessarily fit for drinking

from a chemical perspective and vice versa. Micro biological and chemical aspects

of water quality are important from a health perspective and a water can fail quality

tests in either category.

The most common and widespread danger associated with drinking water is contamination,

either directly or indirectly, by sewage, human or animal excrement, or other wastes.

Fecal pollution of drinking water may introduce a variety of pathogens (bacteria,

protozoa, helminths, etc) into the water, and the use of this water for drinking

or food preparation may cause further cases of infection.

In developing countries, it is usually these infectious water-related diseases which

are of prime importance, not the non-infectious diseases related to some chemical

property of the water.

The recognition that microbial infection can be water-borne has led to simple methods

for the routine examination of drinking water. The micro biological quality of drinking

water is typically expressed in terms of the concentration and frequency of occurrence

of particular species of bacteria.

The detection and enumeration of all pathogens that may be present in water is

far too complex and time consuming to carry out on a routine basis and it is therefore

normal practice to detect and enumerate only what are called "indicator bacteria".

These are bacteria that are always excreted in large numbers by man, whether sick

or healthy.

The presence of indicator bacteria in water is therefore indicative of fecal contamination

of that water. If a sample of water is fecally contaminated, it may contain any pathogen

which is being excreted by the population causing the fecal pollution. Fecal contamination

in a water is thus taken to indicate that water may constitute a health hazard.

The most commonly used indicator bacteria are the coliforms, and water is tested

either for the presence of the total coliform group (chlorinated supplies), or for

the presence of exclusively fecal coliforms (mainly comprising coli) only (unchlorinated

supplies).

The micro biological quality of a water is then expressed as the number of total

coliforms, or fecal coliforms, per 100ml of water. WHO has established guidelines

as to which concentrations should be considered acceptable in given circumstances

(ranging from 0-10 coliform organisms per 100ml depending on the type of supply).

Thus when your article quotes Dr. Chea as saying that "...(the water) contained

a high level of bacteria (between 2.10 and 2.15 mg of organic matter per litre)"

the author has her date mixed up.

Organic load is usually only determined for sewage and waste waters, not for drinking

water. It certainly is no direct indication of bacterial pollution. Besides that,

it is impossible to determine the actual amount of organic matter present in a water.

Organic load is usually expressed as the Oxygen Demand (OD) of a water, in mg per

litre, with a high OD corresponding to a high organic load.

Nothing in your article suggests that micro biological tests were carried out on

the various water samples, which seems to us a serious omission.

To state that "Water drawn from the Tonle Bassac in front of the Royal Palace

also appeared to pass safety tests" is most misleading. Even if bacterial tests

were carried out on the river water, results of such tests should be interpreted

with caution, since turbid waters with many other bacteria present (such as most

tropical rivers) are prone to give false negative results.

Chemical (and physical) water quality standards are commonly laid down for treated

water. For untreated water supplies. chemical water quality standards are generally

inappropriate.

Of most importance from a public health point of view are the inorganic constituents

of a water (organics, such as pesticides, are important as well, but intake in food

is of much more importance here than intake in water).

Tests carried out for the Phnom Penh Post seemed to be focused on the presence of

nitrite.

Nitrate and nitrite should always be considered together, because conversion from

one form to the other occurs in the environment. Concentrations in water are expressed

as mg per litre of nitrate-nitrogen (nitrate-N) and nitrite-nitrogen (nitrite-N).

The first thing to note is that nitrate and nitrite are widespread in the environment;

they are found in most foods, in the atmosphere, and in many water sources.

Levels of nitrate in water are typically below 5 mg of nitrate-N per litre (although

concentrations of 20 times that can occasionally be found) with the levels of nitrite

almost invariably very much lower.

Most of the higher levels of nitrate are found in ground water; in surface water

they tend to get depleted by aquatic plants. Having said that, marked seasonal variations

can occur in concentration in rivers, and high levels may occur especially after

heavy rainfall (eg because of runoff from agricultural areas using fertilizer).

Conventional water treatment and disinfection methods do not affect the levels

of nitrate in a water, but nitrite levels are much reduced through oxidation and

chlorination.

Surface or ground waters which receive organic pollution from sewage discharges or

on-site sanitation systems (such as pit latrines) may show high nitrate levels. A

rising nitrate level in ground water is a warning sign of continuing pollution. Although

boiling water will kill all bacteria, it is likely to increase nitrate concentration

further because of evaporation.

Nitrate concentrations over 20 mg/l in drinking water are potentially hazardous to

health in two ways. The nitrates are reduced in the body to nitrites and can cause

a serious blood condition in bottle-fed infants known as methaemoglobinaemia (also

called infantile cyanosis or "blue baby syndrome"), particularly if the

diet is low in vitamin C.

High nitrate concentrations have also been implicated in the causation of gastric

cancer.

The adverse effects of nitrate invariably involve its reduction to nitrite as a preliminary

step. Hence the ingestion of nitrite leads to a more rapid onset of clinical effects,

and nitrite concentrations should be lower than that of nitrate.

WHO recommends a guideline value of 10 mg/l of nitrate and, although no guideline

value is set for nitrite, it observes that where water is correctly treated, the

nitrite nitrogen level should be considerably lower than 1 mg/l .

The results published in the article are thus incomplete; drawing conclusions from

nitrite levels in a water without considering nitrate levels is useless. It is true

that very high nitrite levels are usually associated with water of unsatisfactory

micro biological quality but the levels of nitrite found are quite low.

Unless Dr Chea Chay has had access to test results that were not reported, classifying

the water as "very toxic" is baseless. It is to be regretted that Dr Chea

Chay overlooked the implications of high nitrate concentrations in water; cases of

serious illness and death in bottle-fed infants are well documented.

All in all, we are afraid that rather than providing new and useful information to

your readers, your article can contribute to unjustified complacency (the river water

passed safety tests) or alarm (certain bottled waters are toxic).

Charged with the responsibility for water supplies in Cambodia, the Department of

Hydrology shares your concern about the water supply situation in the country. [This

month,] a first group of department staff will be trained in carrying out bacteriological

and chemical tests on public water supplies.

Building up a picture of the current situation and trends over the years will

enable the department to formulate sound water development strategies. No one is

served, however, with incomplete or misinterpreted data.

We hope that in the next issue of the Phnom Penh Post you can publish aditional results

justifying your conclusions, or write a clarification based on the background information

in this letter. We would be happy to assist you with interpreting the results of

the tests that were carried out on behalf of the Phnom Penh Post.

Veng Sakhon, vice chief, Water Management Office, Department of Health, and Jan-Willem

Rosenboom, rural water supply advisor (Oxfam).