An interesting analysis of health risk comes from Joshua Cohen and colleagues at the Harvard Center for Risk Analysis (HCRA) and appeared in the American Journal of Preventive Medicine
2005;29(4):325-334. The study examined the public health tradeoffs of an advisory about the risks of methylmercury in fish and shellfish issued by the FDA and EPA in 2004.
Methylmercury is a neurotoxin that can cause developmental delays when young and unborn children are exposed. Fish are an excellent source of lean protein associated with better health and some are rich in omega-3 essential fatty acids. The omega-3 fatty acids are thought to reduce the risk of coronary disease and stroke, as well as promote the neurodevelopment of the fetus and infant.
The advisory was as follows: 1) avoid those types of fish with high mercury concentrations (shark, swordfish, king mackerel and tile fish); 2) eat no more than one meal a week of fish with moderate mercury levels, such as albacore tuna; 3) eat no more than two meals a week of fish with low mercury concentrations (e.g. shrimp, canned light tuna, salmon, pollack and catfish); and 4) check local advisories about the safety of non-commercially caught fish.
Evaluating the tradeoffs of the advisory depends upon the impact on fish consumption patterns, estimating the extent to which the resulting changes in consumption affect nutrient intake and contamination exposure and establishing the relationship between the dietary changes and the resulting health effects. The different types of health effects are aggregated to determine the net impact. One previous study following a similar fish advisory in 2001 found that pregnat women in eastern Massachusettes decreased fish consumption bt 17%. No study reports the impact of advisories on consumption in the broader population.
The HCRA group considered three scenarios for estimating changes in fish consumption resulting from the 2004 advisory. The optimistic scenario
assumes that only women of childbearing age chage their intake of fish by shifting consumption from the typical mix of fish for US consumers to a mix that replaces fish high in mercury with fish low in mercury. This assumes the women reduce mercury exposure while essentially preserving the omega-3 fatty acid benefits and it was the response intended by the FDA and EPA.
In the middle scenario
only women of childbearing age react to the advisory by reducing their consumption of fish by 17% without changing to other fish. This assumes a loss of some neurological development benefits to unborn children due to decreased maternal intake of omega-3 fatty acids. The pessimistic scenario
assumes that all adults reduce fish consumption by 17%, losing some protection against coronary heart disease and stroke.
The analysis includes four health effects: 1) the impact of mercury exposure during pregnancy on cognitive development as measured by IQ; 2) the impact of omega-3 fatty acids on cognitive development measured by IQ; 3) the net impact of fish consumption (i.e., the effect of both omega-3 fatty acids and mercury) on coronary heart disease mortality; and 4) the net impact of fish consumption on stroke-incidence and mortality.
Integrating the fish comsumption scenarios with the estimated impact on health outcomes and aggregating these impacts was done by use of the quality-adjusted life year (QALY). One QALY is defined as a year of life in perfect health, and death is a value of zero QALY. A year of less than perfect health has a value between zero and one QALY.
The impact of the 2004 advisory depends heavily on how it affects fish consumption.
The optimistic scenario
resulted in a gain of more than 400,000 IQ points among the four million babies born annually in the America. There are slight increases in mortality and non-fatal strokes. The cognitive development benefit averages 0.1 IQ points per newborn baby and the increase in individual annual mortality risk is less than one in a million for women between the ages of 35 and 44 years.
Under the middle scenario
the positive effects of omega-3 fatty acids on cognitive development are reduced and the net gain in IQ drops to 92,000 points per year for all newborn babies in the America (approximately 0.02 points per child). The mortality and non-fatal stroke risk remain small.
The pessimistic scenario
results in the same cognitive development benefits to newborn babies because the conditions are identical to the middle scenario for women of childbearing age. The decrease in fish consumption among the adult population results in approximately 8,000 additional deaths and 1,500 non-fatal strokes each year. The risks differ by age group because of the differences in baseline risks. The estimate for a 17% decrease in fish consumption would increase annual mortality risk by approximately 3 per 100,000 for males between 55 and 64 years, by 8 per 100,000 for males between 65 and 74 years and by 28 per 100,000 for males between 75 and 84 years.
Converting IQ points, fatalities and non-fatal strokes into QALY aggregates outcomes and shows whether the overall impact is positive or negative. Under the optimistic scenario
, cognitive development gains among newborns are the dominant factor and population well-being improves. Under the middle scenario
, cognitive development remains the dominant factor, but the countervailing loss due to lower omega-3 intake contributes to reducung the gain by a factor of five. Under the pessimistic scenario
, the cognitive impact for newborns remains beneficial but is dominated by the increased cardiovascular impact among adults.
The analysis found that an increase in fish consumption would decrease both stroke and coronary heart disease risk. Evaluation of epidemiologic data suggests substantial benefits are associated with consumption of at least some fish (e.g., one meal per week) rather than no fish.
The basis of the study and the reason for three scenarios is that the understanding and response to a public health advisory by the population is not uniform, nor does it take the intended course in all cases. Some responses may actually result in greater harm to health. It is an example of the many things that can go wrong and often do.
The questions arise then as to where best to place the emphasis in health and healthcare, how best to educate the public about health, how best to keep the public informed and how best to follow-up on important issues and problems. It begs the question as to the organization and delivery of healthcare in America. The recent posts about projects in Finland, Olmsted county in Minnesota and Greenville county in SC are a beginning. Subsequent posts will continue to address this issue of the organization and operation of healthcare in America.
This information comes from Risk in Perspective
, January 2006, Volume 14, Issue 1.
Technorati Tags: lifestyle