Intended for healthcare professionals

Rapid response to:


Clinicians' roles in management of arsenicosis in Bangladesh: interview study

BMJ 2004; 328 doi: (Published 26 February 2004) Cite this as: BMJ 2004;328:493

Rapid Response:

Arsenic in drinking water in Bangladesh and cancer deaths in 2003 – an environmental health risk assessment

It is excellent that the journal has published on this very serious

It should be emphasised that chronic arsenic poisoning can be
reversed by the provision of arsenic free drinking water. Wells in any
given locality in Bangladesh vary greatly in their arsenic levels, such
that there is often a well with clean water nearby. Reliable testing of
well water, the identification and marking of clean water wells, and the
closure of the most arsenic contaminated wells, should take priority.
Medical identification of people affected by arsenic is only of use if a
safe alternative water supply can be identified. Diagnosing patients with
blackfoot disease (the characteristic pattern of arsenic related
illnesses) or arsenic related cancers may be useful in highlighting areas
for intervention. Perhaps a leaflet on clinical management exists or could
be produced by aid departments or NGOs. Doctors should be involved as
advocates for public health interventions.

Below is a draft abstract of a risk assessment for arsenic and cancer
deaths I hope to have published formally later in the year. Any comments
gratefully received.

Arsenic in drinking water in Bangladesh and cancer deaths in 2003 –
an environmental health risk assessment


Arsenic contamination of drinking water in Bangladesh is likely to
cause large numbers of cancer deaths. Since the 1960s, shallow tube-wells
were introduced widely across Bangladesh to provide drinking water, which
was safe from water borne infectious agents. Wells at depths of 10 to 70
metres struck a shallow aquifer with a high arsenic content. This occurred
across all Bangladesh, but especially in the deltaic South East region.
The features of chronic arsenic poisoning, often known as blackfoot
disease, first became apparent in the 1980s. The British Geological
Survey estimated 35 million people from a 125 million population are
exposed to over 50mcg /L arsenic in drinking water, which has been
associated with a lifetime cancer risk of 1 in 100. This report estimates
the number of cancer deaths, which might occur in Bangladesh in 2003 as a
result of arsenic contamination of drinking water.


An environmental health risk assessment was performed using the WHO
guidelines. The 1998 British Geological Survey in Bangladesh of 2022 wells
(0.03% of all wells in Bangladesh) provided data on arsenic exposure and
its distribution in the population. The health outcomes estimated were
annual totals of deaths from lung, liver, bladder, kidney and all type
cancer. Baseline cancer rates for these cancers for Bangladesh are
available from IARC on the Globocan 2000 database. Data from a study on a
large Taiwanese population exposed to arsenic in drinking water provides
an estimate of the increase in cancer mortality in three categories: 0 –
300mcg/L, 300 – 600 mcg/L and greater than 600mcg/L arsenic in drinking
water. The number of population in each category was estimated, and the
total number of cancer deaths calculated. In view of the controversy
regarding the existence of a safe lower exposure threshold for arsenic
ingestion, total cancer deaths were calculated with and without a 50mcg/L
safe lower threshold.


If there is no threshold for the carcinogenic effects of arsenic, an
estimated additional 35 000 all cancer deaths could result from the
contamination of ground water by arsenic in 2003 in Bangladesh,
representing 28% of cancer deaths in Bangladesh. The overall adult
mortality in Bangladesh in 2000 was estimated by the World Health
Organisation to be 723766 male and female (over 19 years of age) deaths,
indicating that approximately 5% of adult deaths in Bangladesh could
result from the carcinogenic effects of arsenic. Of the cancer deaths, 10
000 would be lung cancer deaths, 1400 would be bladder cancer deaths, 1000
would be kidney cancer deaths, 600 would be from liver cancer. A large
proportion of the all cancer deaths would be from skin cancer. Similar
numbers of male and female deaths would occur, at around 17 000 per year.
If there were a safe no effects threshold at 50 mcg/L, an estimated 30 000
excess cancer deaths might occur.


These results do not include non-cancer arsenic related deaths and
non-fatal disease impacts. Arsenic causes peripheral and cardiac vascular
diseases in addition to cancer. An estimated 220 000 Bangladeshis are
thought to be suffering from arsenic related diseases overall. For
comparison, UNICEF estimates that tube-well use, combined with other
improvements in health care, nutrition and sanitation, have led to a
reduction in child (under five) deaths from 250 000 in 1983 to 110 000 in
1996. Arsenic contamination of groundwater is a public health problem of a
similar order of magnitude to surface water contamination with infectious
agents. Recent grants to ameliorate the problem have not been fully taken
up, possibly because of lack of infrastructure. This needs to be addressed
by aid agencies and governmental departments for international

Competing interests:
None declared

Competing interests: No competing interests

02 March 2004
Ian A. Cameron
Specialist Registrar Public Health
Cotswold & Vale PCT, GL7 1UX