This commentary discusses the issue of how much evidence is enough before recommending in-home interventions aimed at alleviating asthma symptoms.
Department of Public Health and Community Medicine, Tufts University School of Medicine, Boston, Massachusetts.
Correspondence: Doug Brugge, PhD, MS, Department of Public Health and Community Medicine, Tufts University School of Medicine, 136 Harrison Ave, Boston, MA 02111 (email@example.com).
The reviews of in-home interventions to address asthma in this issue of the journal are a valuable contribution to the literature. But they raise the issue of how much evidence is enough before recommending interventions aimed at alleviating asthma symptoms.
A substantial amount of research has accumulated since the Institute of Medicine1 reviewed the evidence for associations of many indoor factors and asthma and allergy. As these reviews show, there is growing and, perhaps in some instances, compelling evidence that some interventions are beneficial to the health of those who receive them. However, there are also some questions that these reviews raise. I will briefly address 2 questions here: (1) What is the appropriate level of proof for environmental interventions such as these and is it justified to apply a lower standard than for other health interventions, such as for medical treatments? Also, a related, but narrower issue, (2) in the absence of truly blinded control or sham intervention groups, should we be concerned about the potential that part or all of the apparent benefit might derive from the social interaction (a Hawthorn or placebo-type effect) rather than from the physical interventions themselves?
Levels of Proof
The Centers for Disease Control and Prevention standard for evidence of efficacy (http://www.thecommunityguide.org/index.htmlmq) is less rigid than the standard used by the Institute of Medicine, which is tied directly to standard epidemiologic frameworks for causal inference, to assess causal associations. It is also less rigorous than the standard applied by the Health Effects Institute in its recent assessment of associations between traffic-related pollution and adverse health outcomes (http://pubs.healtheffects.org/view.php?id=334).
The use of a lower bar to evaluate in-home asthma interventions might be justified. The authors in response to my review of one of their articles argued that these interventions are low risk and unlikely to have unpredicted adverse effects. That could not be said so easily of pharmaceuticals or medical devices. Indeed, for many medical treatments, it is well accepted that there will be substantial adverse effects that are acceptable so long as the benefits outweigh the negatives.
On the other hand, environmental interventions are not without potential to cause harm. The authors point to ozone-generating air filters as a rare example of an intervention that has potential harms. I would argue, perhaps more strongly that these devices are dangerous enough that they should not be on the market. Indeed, the ozone that these devices release is a well-defined toxin with numerous well proven adverse health effects. Secondarily, ozone may also contribute to the formation of toxic particulate matter.2 Thus, there is little question in my mind that at least one potential intervention carries unacceptable risks.
Are ozone-generating devices a unique aberration or are there reasons to be concerned about other interventions as well? I would argue that other in-home interventions might also carry risks. Many of these interventions involve cleaning of one sort or another. Done well, I suspect there is little potential for harm, but done poorly, “cleaning” may make things worse. An example is vacuuming. Use of a properly functioning HEPA filtered vacuum is likely a good idea. Use of a traditional vacuum cleaner or a malfunctioning HEPA vacuum will increase levels of particulate matter in the air, at least in the short term.3
There are other interventions that might have unintended consequences. Pest management that uses few or no neurotoxic pesticides is likely to have positive trade-offs. Use of traditional pesticides might have “hidden” health effects in studies of asthma or allergy because the impact will be in terms of neurological damage and long-term cancer risk, end points that will likely not be measures in asthma studies.4 Chemicals that are toxic to dust mites have been used in numerous studies to control mite populations, with no evidence of efficacy.5
Similarly, there is at least the potential that “green” cleaning products that replace chemicals with known health hazards might at a later date be proven to carry health risks themselves. After all, the screening process for introducing new chemicals into commerce is extremely weak and our ability to predict health effects based on chemical composition without extensive animal testing and epidemiology is limited.
Thus, I would conclude that there are potential risks with in-home interventions aimed at addressing asthma and that such risks argue for appropriate care in evaluating efficacy.
Social Interaction Effects
My second critique is that the social interaction of engaging households might be causing some or all of the beneficial effect in some of the studies cited in support of environmental in-home interventions. The authors of these reviews dismiss this concern largely on the basis of the fact that they think this is unlikely and that, even if it were happening, it would not matter, since these are multifactoral interventions and what matters is that they work when all the combined elements are applied.
Two of the studies that they cite for control of biological exposures6,7 found evidence for a role of social interaction. The first of these studies is the only one that I am aware of that had a control, a sham intervention, and a real intervention. This study found that there was no difference between the sham and real interventions and that both differed from the true control. The Inner City Asthma Study also found substantial improvement in its control group over time, albeit, less improvement than in its intervention group.8 Because other studies have not explored this possibility and these studies found some evidence for a social effect from engaging families with asthma, I am hesitant to dismiss the possibility.
But does it matter? If there is an effect from social engagement and it explains part of the benefit of the interventions, who cares? The intervention is effective, right?
I would argue that it is not that simple and that there are reasons to be concerned if any substantial part of the benefit is not attributable to the reduction of environmental exposures. First, it seems to me that there are 2 possible sources of social effects. The first is that families in the studies are reporting more favorable outcomes, perhaps because they want to believe or perhaps because they want to please the researchers. This is more likely a problem with regard to measures such as self-reported asthma symptom days, a commonly used measure in asthma studies, including the Inner City Asthma Study.8 The second is that they are actually experiencing physiological improvements through psychosocial mechanisms.
If the first case is operating, then we should be very concerned indeed. Under this scenario, there is no real health benefit and the apparent benefit is an artifact. We should be highly motivated to test this hypothesis because our primary interest is improving the health of the target populations. The easiest way to do this is to include sham interventions along with real interventions (and possibly true controls as well). Doing so requires greater resources and effort than most of the studies cited in this issue of the journal, but I would argue that it is worth pursuing. Some interventions will be more amenable to sham interventions (in-home HEPA filters, mite-impermeable bedding, and application of pest control substances) than others (intensive cleaning). Coupling sham interventions with objective markers of health outcomes (eg, emergency department visits or lung function tests) would help address lingering doubts.
In the second case, the benefits are real but not wholly attributable to the environmental intervention(s). The authors have a point here that to some extent this is a circumstance in which we may not care why the intervention is effective. However, I would argue that we would still want, at least over time, to tease apart which elements have little or no contribution to the desired (and real) benefits. By doing so, we could streamline the interventions, potentially reducing their intrusiveness in people's lives and their cost. Furthermore, if social engagement is helpful, we might want to link the environmental intervention to other forms of case management.
I do not want to be construed as arguing against deployment of interventions along the lines of those reviewed in this issue. Rather, I think that while we employ practices based on the best evidence available to date, we should be aiming to better understand our interventions, continue to critically assess them, and seek new versions down the road that are more effective and less likely to have any unanticipated negative consequences. The level of resources devoted to bringing these sorts of interventions to their current state is vastly smaller than the effort to bring most single drugs to market. For asthma, it is critical that we have both in-home environmental interventions and optimized medical care. If we are to have the best possible in-home interventions for health, it may require a level of effort similar to what has been invested in developing pharmaceuticals.
1. Institute of Medicine. Clearing the Air: Asthma and Indoor Air Exposures. Washington, DC: National Academies Press; 2000.
2. Weschler CJ. Ozone's impact on public health: contributions from indoor exposures to ozone and products of ozone-initiated chemistry. Environ Health Perspect. 2006; 114:1489–1496.
3. Corsi R, Siegel JA, Chiang C. Particle resuspension during the use of vacuum cleaners on residential carpet. J Occup Environ Hyg. 2008; 5:232–238.
4. Julien R, Levy JI, Adamkiewicz G, et al. Pesticides in urban multiunit dwellings: hazard identification using classification and regression tree (CART) analysis. J Air Waste Manage Assoc. 2008; 58:1297–1302.
5. Gøtzsche PC, Johansen HK. House Dust Mite Control Measures for Asthma. Allergy. 63(6):646–659.
6. Carter M, Perzanowski MS, Raymond A, et al. Home interventions in the treatment of asthma among inner-city children. J Allergy Clin Immunol. 2001; 108:732–737.
7. Levy JI, Brugge D, Peters JL, Clougherty JE, Saddler S. A community-based participatory research study of the efficacy of multifaceted in-home environmental interventions for pediatric asthmatics in public housing. Soc Sci Med. 2006; 63:2191–2203.
8. Morgan WJ, Crain EF, Gruchalla RS, et al. Results of a home-based environmental intervention among urban children with asthma. N Engl J Med. 2004; 351:1068–1080.