Notes Washed Up in a Bottle

Don. W. MacCorquodale M.D. M.S.P.H.

Periodic Notes From the Field on Philosophy and Science.

AMBIENT OZONE AND HEALTH  

          In 2003 the US Environmental Protection Agency funded three separate research teams to address the question of whether or not the epidemiological literature suggested that ozone exposure resulted in increased risk of premature death (The editors.   On the health effects of ozone exposure and exposing the epidemiologic process.   Epidemiology 2005; 16:   425-426).   The three groups received identical databases, and they conducted their analyses independently.   The journal Epidemiology received independent submissions from the three teams.   Each was a meta-analysis of published studies on ozone and mortality.   All three studies were published in Epidemiology (Bell M L et al.   A meta-analysis of time series studies of ozone and mortality . . . 436-445; Ito K et al.   Association between ozone and daily mortality . . . 446-457; Levy J L et al.   Ozone exposure and mortality . . . 458-468).  

          The findings of the three studies were succinctly summarized by David V. Bates (Ambient ozone and mortality.   Epidemiology 2005; 16:   427-429).   All three studies showed a “small but substantial association between ozone levels and total mortality.” The estimates of mortality effects were remarkably consistent within a fraction of a percent – “an 0.8% increase in immediate mortality per 10-ppb increase in average daily ozone over the year.”   The three teams used different but overlapping databases.   Bayesian hierarchical models were employed in all three analyses.   PM (particulate matter) interaction with ozone did not prove to be important.   All three studies found that the effects of ozone exposure were higher in summer when ozone levels are higher.  

          The question of personal exposures was complex.   Nitric oxide, NO, actually “quenches” ozone to form nitrogen dioxide, NO 2 .   It is possible that people living in the inner city are personally exposed to lower ozone levels than wealthier people in the suburbs.  

          Another important factor is the time of ozone formation, which usually reaches a maximum between 12 noon and 3:00 p. m.   Children in air-conditioned schools are exposed to indoor ozone levels that are only about 15% of ambient outdoor levels.  

          Is it biologically plausible that ozone exposure could increase mortality?   Dr. Bates suggested that might be the easiest question to answer.   “Ozone is capable of causing inflammation in the lungs at lower concentrations than any other gas.”  

          Dr. Bates concluded that these three studies “point to the urgent need to reduce public exposures to ozone by all possible means.”

          COMMENT:   Some years ago while I was working for the US Army Center for Health Promotion and Preventive Medicine, I reviewed some 50-60 articles on the health effects of air pollution, and I was struck by how very different the findings were from studies conducted in a variety of cities at home and abroad.   All showed health effects, primarily increases in visits to hospital emergency rooms for respiratory problems, but there were remarkable dissimilarities.   One study showed the culprit was ozone; the next, sulfur dioxide.   I seem to remember that the most widely accused culprit was PM 2.5 .   One study showed that health effects appeared the same day as the increase in pollution.   The next study showed the health effects appeared with a lag of one day; still another, with a lag of two days.

          So, it is hardly remarkable that I was very impressed that all three research groups in this instance produced similar findings and conclusions.

The EPA is to be congratulated for funding these studies.    

ORGANOCHLORINES AND NON-HODGKIN LYMPHOMA

          The incidence of non-Hodgkin lymphoma (NHL) has increased appreciably in a number of countries during the last few decades (Colt J S et al.   Organochlorines in carpet dust and non-Hodgkin lymphoma.   Epidemiology 2005; 16:   516-525).   Such compounds as DDT and PCBs are widespread environmental pollutants that accumulate in adipose tissue, and several case-control studies had found an elevated risk of NHL associated with exposure to DDT or chlordane.   Other studies have suggested that risk of NHL is related to occupational exposure to pentachlorophenol and other chlorophenates.   Such studies are difficult to interpret because the commercial grades of these substances may be contaminated with dioxins and dibenzofurans.  

          The authors conducted a population-based case-control study of adult NHL.   Carpet dust samples were collected and analyzed for organochlorides, PCBs, and other substances.   Their article examined the association between organochlorine concentrations in dust and NHL risk.  

          Cases were newly diagnosed men and women age 20-24 years with newly diagnosed primary NHL in four areas covered by the Surveillance, Epidemiology, and End Results Program of the National Cancer Institute.   Controls were selected from the general population by random digit dialing, and they were frequency-matched on age, sex, and race.   Personal interviews were conducted in participant's homes, and each subject was asked for permission to collect a dust sample.   Those who consented were screened for eligibility.  

          Of slightly more than 1,400 eligible cases, 187 were deceased, could not be located, or had physician refusal.   The authors attempted to contact the remaining 1,100 (app.), and they were successful in interviewing 89% of those contacted.   Of about 1,500 eligible controls, the authors attempted to contact 1,400 (89%), and they were able to interview slightly more than 800 (60%).  

          Multiple logistic regression models were used to estimate the relative risk of developing NHL by deriving odds ratios while controlling for age, sex, and education.   There was a 30% elevation in NHL risk if DDE was detected in the dust (95% CI:   1.0-1.7).   A 50% excess was observed if any PCB congeners were detected (95% CI:   1.2-2.0).   There was no evidence of an association with DDT.

          Previous studies have noted increased risk in individuals with higher concentrations of PCBs in blood or adipose tissue.   In a nested case-control study of adipose tissue samples collected from cadavers and surgical patients between 1969 and 1981, no association between PCB levels and NHL risk was noted.   Earlier epidemiologic studies have usually found modest elevations of NHL risk in association with DDT, although no association was found in a population-based case-control study in Sweden .  

          The authors pointed out that one of the strengths of their study was the large number of participants with dust samples.   A weakness was the loss of information from death, inability to locate, etc.   If organochlorine exposure shortens survival, they may have underestimated risk.  

          COMMENT:   I find it somewhat surprising that so little is know about the epidemiology of NHL when the incidence of the disease has been increasing so strikingly, particularly in this country.   Unhappily, this study does not provide compelling evidence of an association between organochlorines and the disease.

CHILDHOOD LEUKEMIA AND SOCIOECONOMIC STATUS       

          Previous studies of childhood leukemia have demonstrated higher incidence in association with higher socioeconomic status (SES).   More recent case-control studies of low-frequency magnetic fields and childhood leukemia found cases tended to be of lower SES (Borugian M J et al.   Childhood leukemia and socioeconomic status in Canada .   Epidemiology 2005; 16:   526-531).   If SES is a risk factor for childhood leukemia, then, SES could also behave as a confounder.   The authors analyzed age-standardized incident rates of all childhood leukemia cases throughout Canada in relation to SES using population-based quintiles of neighborhood income adjusted for household size.  

          A higher incidence of leukemia was seen among younger children with a peak in the 0- to 4-year age group.   Incidence was slightly higher in boys.   Slightly more than three fourths of the cases (77%) were acute lymphoid leukemia.   For children diagnosed with acute lymphoid leukemia, the rate ratio for the poorest neighborhood income quintile compared to the richest was 0.86, 95% CI:   0.78-0.95.

          COMMENT:   Upper SES has long been regarded as a risk factor for childhood leukemia.   A number of observers have suggested that since exposure to ionizing radiation, including x-rays, is a function of social class in children, it is responsible for the increased risk in children of lower social class.   This study has the advantage of an enormous “sample,” and I confess to a certain amount of surprise that the effect is relatively small.