From the Institute for Risk Assessment Sciences, Utrecht University, Utrecht, The Netherlands.
Address correspondence to: Bert Brunekreef, Institute for Risk Assessment Sciences, Utrecht University, P.O. Box 80176, 3508 TD Utrecht, The Netherlands; B.Brunekreef@iras.uu.nl
When one enters my hometown of Utrecht, the Netherlands, from the east, there are several signs in Polish explaining the dos and don’ts of the large used-car market held there every Tuesday. The signs are there because, ever since the Berlin Wall came down in 1989, most of the market’s clients have been coming from Eastern Europe (notably Poland). Large numbers of old cars, not-so-well maintained and therefore not-so-clean, have found their way east, contributing to mobility and new forms of air pollution throughout Central and Eastern Europe. Other changes have improved the environment. Many heavy industries collapsed because they could not compete in the open market, and home heating systems have been overhauled, leading to a substantial reduction in emissions of “traditional” air pollutants such as sulfur dioxide (SO2) and total suspended particulate matter (TSP).
In this issue, Heinrich and colleagues present an intriguing study of the consequences of political change on public health. 1 The investigation shows the decrease in SO2 and TSP over a period of 7 years (1992–1999) after German reunification, and with this, the decline in prevalence of bronchitis and colds. This observation extends a previous report, spanning 1992–1996, on the same population of children. 2
The Heinrich study adds to a small but growing set of studies on possible effects of changes in air pollution over time. The time scale here (years) is very different from the changes in air pollution exposure over hours to days described by extensive time-series studies. Literally hundreds of such short-term studies have been conducted over the past 15 years, using routinely collected data on air pollution concentrations and daily mortality or hospital admissions.
Why does the time scale matter? In time-series studies, the focus is on short-term variation of exposure primarily driven by variations in weather. The number of potential confounders is limited; everything that changes over a time period less than a couple of weeks is filtered out by smoothing techniques or by the application of linear or nonlinear trend adjustments. As a result, these time-series studies do not capture the effects of slower changes. In contrast, the Heinrich investigation 1 specifically addresses long-term trends in exposure and outcome. Clearly, if long-term reductions in air pollution can be shown to have public health benefits, this would support further efforts to regulate air pollution.
“. . .Heinrich and colleagues present an intriguing study of the consequences of political change on public health.”
Previous cross-sectional studies on effects of air pollution in children have documented associations between various air pollution components and bronchitis rates. 3 However, there have generally been no clear associations with wheeze or asthma. 4 In keeping with the cross-sectional findings, the studies from eastern Germany find decreases in doctor-diagnosed bronchitis and the occurrence of frequent colds in the past 12 months (all self-reported), but virtually no decrease in current wheeze.
Other studies conducted by the same team have even suggested that the prevalence of allergic disease has increased in the eastern part of Germany after the reunification. 5,6 It has been difficult to show convincingly that increases in the reported prevalence of allergic disease were not caused at least in part by a change in diagnostic habits of physicians and a change in awareness among study subjects, 6 although this change would not explain the observed increases in sensitization to common allergens.
Similarly, one could argue that what was previously diagnosed as “bronchitis” may to some extent now be labeled as “asthma,” leading to a decrease in one and an increase in the other. There is a slight indication of this in the Heinrich study, in that, within the cohort, the prevalence of “ever having bronchitis” decreased with time (Table 4), which it theoretically cannot do. However, another explanation might be that when children get older, parents forget about diagnoses of relatively minor bronchitis episodes that they may have reported in an earlier questionnaire. Regardless, in the near future one would hope to see a joint analysis of the several excellent German “reunification” studies provide more insight into this difficult question of diagnostic shifts over time.
Another question that comes to mind is what this study tells us about “safe” levels of exposure. TSP and SO2 annual means were still high in the early 1990s, as is apparent from Table 1. 1 Most of the change in TSP occurred in the early part of the study, and this was also true for most of the decline in reported bronchitis. SO2 continued to decrease in the latter part of the study, although this decrease seems not to have been associated with further declines in bronchitis.
Age-specific effects may also be an issue. The air pollution situation in the former East Germany was very dynamic in the period immediately preceding this study. The three oldest cohorts were born in the early 1980s (Table 4) and exposed to a series of air pollution episodes later in the 1980s. A particularly severe episode occurred in January 1985. 7,8 It lasted 16 days, and maximum 24-hour average SO2 concentrations in nearby Leipzig reached an astonishing 2290 μg/m3. This episode covered large portions of Western Europe as well, where it triggered several investigations. 9–11 These investigations, in turn, revived interest in the health effects of air pollution, which just a few years earlier had been declared to have become trivial. 12 Pollution episodes in January 1987 and November/December 1989 were characterized by maximum 24-hour average SO2 concentrations of 1250 and 1500 μg/m3, respectively. Annual averages in Leipzig were between 196 and 377 μg/m3 through the 1981–1989 period. It is possible that the more recent difference in bronchitis rates between the older and younger cohorts in the Heinrich study may in part reflect the lingering effects of earlier exposures.
What further evidence do we have from similar “natural experiments”? Obviously, air pollution abatement measures taken in the U.K. in the wake of the disastrous killer smog episodes of 1952 provided strong evidence that the extremely high pollution levels observed in London (and elsewhere) before the mid-1960s were having serious detrimental effects on public health. Pollution levels such as those reported by Heinrich in the former East Germany in the 1990s were, however, only a fraction of those found in the London smogs of the 1950s, and with less dramatic effects. One would like to see support from studies conducted in other areas before becoming more strongly convinced that further reductions in long-term average air pollution exposures are indeed beneficial.
Fortunately, such studies are becoming available. One important example is the study by Pope 13 that documents fewer hospital admissions for respiratory disease in Utah Valley during a period when the major local source of air pollution (a steel mill) was shut down by a strike. This study also suggested that fewer people died during the low-pollution period. Recently, filters containing airborne particulate matter collected before, during, and after the strike were located, and the collected particles were analyzed for toxicity. The particles collected during the low-pollution period of the strike were much less toxic, even adjusting for weight, than particles collected before and after the strike. 14,15 This study is one of the first examples in which an epidemiologic observation of a beneficial effect of air pollution reduction has been corroborated by toxicologic studies.
We should not be naive about how much such a study “proves” that lower air pollution improves public health. Changes in exposure such as those studied by Pope 13 and Heinrich 1,2 never occur in isolation. In the case of the German reunification, just about everything changed—not just air pollution exposures. To disentangle the effects of declining air pollution in reunified Germany from changes in lifestyle, personal habits, and medical care is a challenge. In the Utah Valley example, the change in air pollution was a more isolated event, but even there, changes in insurance status related to the strike may have affected the willingness to seek medical care.
In regulatory circles, more and more questions are being asked about the possible benefits of further air pollution abatement. How much will public health improve as we start “sliding down the dose-response slope”? Randomized community trials of the effects of air pollution abatement are clearly not feasible. To answer these important dose-response questions, we need more rigorous frameworks for interpreting the outcomes of “natural experiments,” such as the reunification of Germany. In the absence of true experimental studies, the importance of studying environmental changes that spring from political and societal events will only increase.
1. Heinrich J, Hoelscher B, Frye C, et al. Improved air quality in reunified Germany and decreases in respiratory symptoms. Epidemiology 2002; 13: 394–401.
2. Heinrich J, Hoelscher B, Wichmann HE. Decline of ambient air pollution and respiratory symptoms in children. Am J Respir Crit Care Med 2000; 161: 1930–1936.
3. Dockery DW, Cunningham J, Damokosh AI, et al. Health effects of acid aerosols on North American children: respiratory symptoms. Environ Health Perspect 1996; 104: 500–505.
4. Kaur B, Anderson HR, Austin J, et al. Prevalence of asthma symptoms, diagnosis, and treatment in 12–14 year old children across Great Britain (international study of asthma and allergies in childhood, ISAAC UK). BMJ 1998; 316: 118–124.
5. Heinrich J, Richter K, Magnussen H, Wichmann HE. Is the prevalence of atopic diseases in East and West Germany already converging? Eur J Epidemiol 1998; 14: 239–245.
6. Heinrich J, Hoelscher B, Frye C, Meyer I, Wjst M, Wichmann H-E. Trends in prevalence of atopic diseases and allergic sensitisation in children in eastern Germany. Eur Resp J 2002; (accepted for publication).
7. Werner H, Berger F. Wintersmog in der ehemaligen DDR Teil I: Ursachen, Kriterien und Datengewinnung. (Winter air pollution in the former GDR part 1: criteria and data acquisition.) Staub 1992; 52: 119–126.
8. Werner H. Wintersmog in der ehemaligen DDR TeillI: Datenanalyse, Diskussion und Ausblick. (Winter air pollution in the former GDR part 2: data analysis, discussion and perspectives.) Staub 1992; 52: 239–244.
9. Dassen W, Brunekreef B, Hoek G, et al. Decline in children’s pulmonary function during an air pollution episode. J Air Pollut Control Assoc 1986; 36: 1223–1227.
10. Ayres J, Fleming D, Williams M, McInnes G. Measurement of respiratory morbidity in general practice in the United Kingdom during the acid transport event of January 1985. Environ Health Perspect 1989; 79: 83–88.
11. Wichmann HE, Mueller W, Allhoff P, et al. Health effects during a smog episode in West Germany in 1985. Environ Health Perspect 1989; 79: 89–99.
12. Holland WW, Bennett AE, Cameron IR, et al. Health effects of particulate pollution: reappraising the evidence. Am J Epidemiol 1979; 110: 527–659.
13. Pope CA 3rd. Respiratory disease associated with community air pollution and a steel mill, Utah Valley. Am J Public Health 1989; 79: 623–628.
14. Dye JA, Lehmann JR, McGee JK, et al. Acute pulmonary toxicity of particulate matter filter extracts in rats: coherence with epidemiologic studies in Utah Valley residents. Environ Health Perspect 2001; 109 (suppl 3): 395–403.
15. Ghio AJ, Devlin RB. Inflammatory lung injury after bronchial instillation of air pollution particles. Am J Respir Crit Care Med 2001; 164: 704–708.
© 2002 Lippincott Williams & Wilkins, Inc.