Cartas / Letters


La Revista Panamericana de Salud Pública/Pan American Journal of Public Health se complace en publicar cartas de los lectores dirigidas a estimular el diálogo sobre los diversos aspectos de la salud pública en las Américas, así como a esclarecer, discutir o comentar de manera constructiva las ideas expuestas en la revista. Las cartas deben estar firmadas por el autor y especificar su afiliación profesional y dirección postal. Cuando se trate de comentarios sobre un artículo que requieran contestación del autor, se procurará conseguir esa respuesta con el fin de publicar ambas cartas. La Redacción se reserva el derecho de editar las cartas recibidas y resumirlas para mayor claridad.

The Revista Panamericana de Salud Pública/Pan American Journal of Public Health publishes letters from readers for the purpose of stimulating dialogue on various aspects of public health in the Americas and of constructively clarifying, discussing, and critiquing the ideas expressed throughout its pages. Letters should be signed by the author and include his or her professional affiliation and mailing address. If a commentary on a given article requires a reply from the author, an effort will be made to obtain the reply and to publish both letters. The editorial team reserves the right to edit all letters received and to condense them so as to improve their clarity.



What quality standards are appropriate for rural water supplies?



The paper by Welch et al. entitled "Microbial quality of water in rural communities of Trinidad" (1) discussed aspects of water quality testing and expectations of safety of rural water supplies, and it affirmed similar observations of unpotability in developing countries that my colleagues and I described in a study in sub-Saharan Africa (2). The hazard of increased contamination of stored water supplies in the home must be emphasized (3).

Consumers expect that delivered water, whether by truck or pipe, and that is reported to be sanitized, should be free of total coliforms and Escherichia coli. That expectation is in line with guidelines published by the World Health Organization (4). Finding concentrations of these indicator bacteria in treated water supplies implies that treatment procedures are inadequate, quality control measures have lapsed, or a breach in the integrity of the delivery or distribution system has occurred after treatment, as well as that there is a possibility of postdelivery contamination.

Respectfully, I suggest that several assumptions by the authors beyond this observation miss the mark. Applying in rural areas the potability standards of the United States Environmental Protection Agency (EPA)¾that is, an absence of microbial indicators¾is a misleading, untenable, unrealistic, and unnecessary goal for water delivery schemes in many developing countries. Providing drinking water free of coliforms is an admirable and idealistic aspiration. However, EPA guidelines, while appropriate in locations with adequate resources, ought not to be applied to civil infrastructure in geographical settings where neither technical expertise nor resources are available to guarantee delivery of safe drinking water. In addition, the authors do not place in perspective the reality of developing countries, where people may not understand concepts of pathogenicity and most rural water "systems" rely on surface sources or shallow wells.

Our research, conducted in Lesotho in the early 1990s, was able to distinguish potability markers and engineered solutions that practically and realistically indicated whether a water supply was likely to be safe or to be potentially hazardous. Gravity-feed water system configurations that had been implemented by development projects in previous years generally provided water considered safe for human consumption if they were properly maintained. Total coliforms were present, while E. coli were usually absent. These systems had a protected intake at the surface source, connected by plastic pipe first to a sediment tank and then to either a plastic or corrugated metal storage tank with a tap. This clever construct required inexpensive and minimal technical knowledge to implement. These systems did not rely on chlorination or other mechanisms of disinfection to attain potability. The goal of the design was to protect the water source from human and/or animal fecal contamination

Ubiquitous in the environment, total coliforms represent a taxonomically diverse classification of bacteria, including E. coli. The source of coliform bacteria may be from fecal sources or biodegraded vegetable matter, with the latter not considered to be harmful or of sanitary significance to humans. To infer that water supplies are unsafe if coliforms as a class are present is an inaccurate application of water quality bacterial indicator screening tools. A more accurate screening tool, based specifically on the presence of E. coli, is the only logical indicator of potability in settings where disinfection is not a realistic option.

Some would undoubtedly argue that ignoring the public health needs of rural, disadvantaged villagers by not "protecting" their drinking water from coliforms is an unacceptable position and is based on a double standard. But, given the realities in many parts of the world, the aspiration of protecting water supplies through expensive, technical interventions is in large measure impossible and excessive. There is always a danger that an inability to comply with an unreasonable standard may paralyze any remedial efforts. Tempered goals are required. Implementing rational solutions based on local resources and capabilities is recommended (5). There need not be an expectation that water supplies are "sterile." Protecting a water source from fecal contamination is the key to a practical potability standard based on bacterial indicator quality control measures.



1. Welch P, David J, Clarke W, Trinidade A, Penner D, Bernstein S, et al. Microbial quality of water in rural communities of Trinidad. Rev Panam Salud Publica 2000;8(3): 172-180.

2. Kravitz JD, Nyaphisi M, Mandel R, Petersen E. Quantitative bacterial examination of domestic water supplies in the Lesotho Highlands: water quality, sanitation, and village health. Bull World Health Organ 1999;77(10):829-836.

3. Molbak K, Hojlyng N, Jepsen S, Gaarslev K. Bacterial contamination of stored water and stored food: a potential source of diarrhoeal disease in West Africa. Epidemiol Infect 1989;102(2):309-316.

4. Guidelines for drinking-water quality. Volume 1: recommendations. 2nd ed. Geneva: World Health Organization; 1993.

5. Morgan P. Rural water supplies and sanitation. Hong Kong: MacMillan Education Ltd.; 1990.


Jay D. Kravitz
Oregon Health Sciences University
Department of Public Health
and Preventive Medicine
Portland, Oregon, United States of America

Organización Panamericana de la Salud Washington - Washington - United States