Report No DWI4304/2

Oct 1997

The European Commission has proposed a revision of the Drinking Water Directive; for lead, it is proposed to reduce the limit from 50 to 25 µg/l for an interim period, with an ultimate limit of 10 µg/l to be achieved within 15 years of the implementation of the revised Directive. These limits are to apply to "representative" water samples although the precise form of the standard (e.g. mean, maximum) has not yet been determined. Even with the phasing-in of the new limit over a period of several years, this will incur considerable costs for water undertakers (for water treatment and communication pipe replacement) and property owners (supply pipe and plumbing replacement).

Lead in drinking water has long been associated with detrimental impacts on consumers' health status. Changes in health status which have been examined have included the impact on IQ and lead-induced hypertension effects. It is possible to attach monetary values to the improvements in health status which would result from reducing exposure to lead from drinking water. A programme to reduce lead exposure would also generate other benefits and costs such as reduced leakage from a modernised distribution system and the disruption this modernisation would cause.

A previous project ("the Cost Study") determined the costs to water suppliers (water treatment and pipe replacement) and householders (pipe replacement) of compliance with the various possible interpretations of the interim and final standards and monitoring requirements and compared these qualitatively.

The objective of this project is to carry out a cost benefit analysis utilising the cost information provided by the Cost Study in order to assess the economic benefits likely to arise from meeting various possible interpretations of the interim and final standards.

Two policy options were considered: compliance with the revised Directive and the "do nothing scenario". In addition the study assessed the benefits arising from meeting the proposed interim standard of 25 µg/l as a maximum and as an average and the proposed final standard of 10 µg/l as a maximum and as an average. The cost information was taken directly from the Cost Study.

Relationships were established between lead in water and associated health effects such as the effect on childhood IQ. Dose response relationships were examined in terms of the exposure-damage relationships between lead levels, the numbers of people affected and the ways in which their health and development is impacted. Finally the benefits were estimated, where possible, in quantitative terms.

From the scientific literature, it is possible to derive models relating relevant low levels of blood lead to IQ and blood pressure. These models were used in the cost benefit analyses. Other endpoints which have been associated with low blood lead levels, such as effects on gestational age and biochemical indices, were not taken into account. Appropriate quantitative models for these endpoints are less well developed, and the clinical significance of, for example, minor changes in biochemical indices, is unclear.

Relationships were developed between waterborne lead and blood lead levels, with adjustments to account for the fall in blood lead levels since the original models were developed. However, it is likely that orthophosphate treatment of drinking water to reduce lead solubility will complex dissolved lead ions and reduce uptake; this could have a significant impact on the relationship between water lead and blood lead concentrations.

Current exposure to water lead was derived from the results of water company compliance sampling; future exposure was estimated for the "do nothing" and interim and final standard cases interpreted as an average and a maximum. Populations affected were derived from official actuarial figures.

The benefits that would accrue through increased IQ on future earnings were monetarised, taking into account the several linkages between lead, IQ, school performance, work-force participation and wages. The economic benefits associated with a reduction in blood pressure were estimated based on the linkages between lead, hypertension and morbidity and mortality, using a quality adjusted life years (QALYs) approach and valuation of quality and length of life changes. Non-health benefits and disbenefits were assessed: it was possible to value some of these effects; however, their monetary values were small compared to the value of the health related benefits.

Given the uncertainty associated with the estimates, the overall appraisal considers 'best case' and 'worst case' scenarios. Best case results are estimated on the basis of the highest benefits and the lowest costs derived from the separate evaluations of impact, whilst worst case results are estimated on the basis of the lowest benefits and the highest costs.

All costs and benefits are discounted at a 6% real interest rate over the period 1998 to 2034(1) and are presented below in 1995 prices. In addition sensitivity tests were conducted on 4% and 2% discount rates. Costs are shown as negative, and benefits as positive, values; the net benefits are calculated as the sum of the (negative) costs and benefits. These figures are for England and Wales.

(l)A terminal year of 2034 is used as it is the last year for the population forecasts used in the study and sufficiently far in the future to capture all relevant costs and benefits. Sensitivity analyses have confirmed that going further than 2034 does not alter the outcome.

Copies of this report may be available as an Acrobat pdf download under the 'Find Completed Research' heading on the DWI website.