Review of Existing Literature on Quantifying and Valuing Human Health Risks Associated with Low Level Exposure to PCBs

1.4 Health risk conclusions

The intent of this review was to provide low, central, and high estimates of the concentration-response relationships suitable for a DFA- based benefit analysis. However, it is not possible to do so from the published reviews of the PCB health effect literature. The overall conclusion of this report is that although there is a great deal of scientific information available concerning the human health risks from exposure to PCBs, including considerable information on low-level or chronic human exposures from environmental exposure, there is insufficient scientific information available at this time to identify a concentration-response function for any of the known or suspected PCB-related health risks.

Table 1 summarizes the conclusions about the availability of PCB-related concentration-response functions for the four major health effects of concern from low-level chronic PCB exposure: immune system deficiencies, neurological impairment, developmental impairment, and cancers.

Further follow-up is not recommended for PCB-related immunotoxicity and developmental impairment because the evidence for an effect from environmental exposures is weak or equivocal. Follow-up may not be needed for cancer. While PCB toxicity, persistence, and environmental bioaccumulation processes suggest all PCB mixtures may cause cancer (see Section 2.4 on cancer), the evidence linking low-level PCB exposures and cancers is inconclusive. One population of concern are particular subgroups of women who may be at increased risk for breast cancer. Environmental PCB exposures do appear to contribute to neurological impairment, and further review of the individual publications theoretically could identify specific damage functions. However, clear estimates may still not be possible because PCB exposures differed across geographic and temporal scales, what was measured differed across studies, and the different PCB mixtures have different levels of toxicity, making determination of specific concentration-response relationships difficult.

Table 1. Summary of concentration-response conclusions

Health effect group	Sources of PCB exposure	Specific health effects	Conclusions
Immune system deficiencies	Infant exposure: in utero and breast milk Adult exposure: fish consumption and general environmental exposures	Infectious illnesses, ear infections, immunological parameters	Studies suggest human system sensitivity. However human studies provide only limited evidence of PCB immunotoxicity due to mixed chemical nature of the exposures and generally insufficient information on exposure-response relationships.
Neurological impairment	Infant exposure: in utero and breast milk	Infant visual and muscular effects; older child verbal, cognitive, motor skills and behavioral effects	Great concern even low levels of in utero PCB exposures may produce long-lasting neurological damage. However, due to the limitations of epidemiological studies, these effects cannot be attributed entirely to PCB exposure.
Developmental impairment	Infant exposure: in utero and breast milk	Reduced reflexes, low muscle tone, low birth weight	No firm conclusions can be made regarding growth and development of children and environmental exposures to PCBs, although perinatal exposure to high concentrations of PCBs affects birth weight and growth during early life.
Cancers, fatal and non-fatal	Occupational exposure: more limited information on environmental exposure	Cancers of the liver, biliary tract, intestines and skin. Possible breast cancer and non-Hodgkin's lymphoma	Human studies provide some evidence that PCBs are carcinogenic, especially for occupational exposures. General population studies are inconclusive with respect to associations between environmental exposures to PCBs and risk of breast cancer or non-Hodgkin's lymphoma. There are preliminary indications that particular subgroups of women may be at increased risk for breast cancer.

1.5 Valuation conclusions

There are two primary approaches to establishing a value (or valuation function) for adverse health effects needed in a DFA-based environmental benefit assessment. The approach most readily understood by the general public uses financial-based measures of the actual cost of medical treatment for the health effect (Cost of Illness), as well as other illness-related financial costs such as lost income for the patient and family members involved with the care of the patient. A broader valuation concept that is preferred by economists uses more inclusive measures that reflect the full consequences to the affected individuals and to society. The broader valuation concepts can include quality of life impacts, as well as the financial-based costs of illness. In applied benefits analysis, economists use valuations based on the more inclusive measures for as many endpoints as possible. For health effects for which inclusive valuation measures are not available, economists use valuations based on financial measures, which underestimate the comprehensive value.

Valuation information of any sort is only available for a subset of the known and suspected health effects associated with exposure to PCBs. Valuation studies for many of the health effects of concern, including such serious effects such as reproductive disruption, most neurological damage, and weakening of the immune system, have not been successfully conducted. The recommended valuations for a limited set of PCB-related health effects, including the health effects of primary concern as well as potential health effects, are summarized in Table 2. All of the values in Table 2 are in Canadian dollars, reflecting Canadian 2005 prices and incomes.

1.6 Organization of the remainder of this report

Section 2 provides a technical summary of the available information on health risks from PCBs. Section 3 introduces the concepts of health effect valuations used in a DFA, and presents the available valuation information for health effects associated with PCBs. Section 3 also discusses the calculations involved in presenting the valuations in terms of 2005 Canadian dollars. The final section provides the references used in this report.

Table 2. Summary of recommended valuations^a

Health effect	Central estimate	Low estimate	High estimate	Relative confidence
Premature mortality	$7.5 million from Canadian studies (alternative: $7.1 million from multi-national studies)	$3.6 million	$9.9 million	High: based on multiple reviews of hundreds of Value of Statistical Life studies worldwide, used in many regulatory settings with extensive review. However, important issues remain uncertain, including role of specific type of risk, expected loss of life expectancy, etc.
Liver cancer	$247,000	$185,000	$370,000	Medium: based on older COI study. High estimate based on Willingness to Pay/COI adjustment introduces additional uncertainty.
Non-Hodgkin's lymphoma	$155,000	$90,000	$186,000	High: based on recent, high quality COI study. High estimate based on WTP/COI adjustment introduces additional uncertainty.
Hypertension	$16,000	$16,000	$32,000	Medium: based on older COI and lost income study.
Coronary heart disease	$95,000	$95,000	$140,000	Medium: based on older COI and lost income study.
Stroke	$305,000	$305,000	$460,000	Medium: based on older COI and lost income study.
IQ point loss	$12,500	$15,000	$25,000	Low: based on older effect on lifetime earnings study. Extrapolation to Canada less certain than for other health effects.
Acute bronchitis	$500	$85	$900	Medium: based on modern valuation study.
Upper respiratory symptoms	$35	$15	$65	Low: based on older valuation study.
Lower respiratory symptoms	$23	$10	$35	Low: based on older valuation study.
Asthma exacerbation	$60	$25	$100	Medium: based on both older and modern valuation study.
Minor restricted activity days	$70	$30	$120	Medium: based on both older and modern valuation study.
a. All values in 2005 Canadian dollars.